Automatic Speech Recognition
Transformers
ternary-quant
quantization
ternary
audio
speech-to-text
whisper
Instructions to use AsadIsmail/whisper-large-v3-ternary with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AsadIsmail/whisper-large-v3-ternary with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="AsadIsmail/whisper-large-v3-ternary")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AsadIsmail/whisper-large-v3-ternary", device_map="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "model_name": "openai/whisper-large-v3", | |
| "model_config": { | |
| "transformers_version": "5.5.3", | |
| "architectures": [ | |
| "WhisperForConditionalGeneration" | |
| ], | |
| "output_hidden_states": false, | |
| "return_dict": true, | |
| "dtype": "float32", | |
| "chunk_size_feed_forward": 0, | |
| "is_encoder_decoder": true, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "problem_type": null, | |
| "vocab_size": 51866, | |
| "num_mel_bins": 128, | |
| "encoder_layers": 32, | |
| "encoder_attention_heads": 20, | |
| "decoder_layers": 32, | |
| "decoder_attention_heads": 20, | |
| "decoder_ffn_dim": 5120, | |
| "encoder_ffn_dim": 5120, | |
| "encoder_layerdrop": 0.0, | |
| "decoder_layerdrop": 0.0, | |
| "decoder_start_token_id": 50258, | |
| "use_cache": true, | |
| "activation_function": "gelu", | |
| "d_model": 1280, | |
| "dropout": 0.0, | |
| "attention_dropout": 0.0, | |
| "activation_dropout": 0.0, | |
| "init_std": 0.02, | |
| "scale_embedding": false, | |
| "max_source_positions": 1500, | |
| "max_target_positions": 448, | |
| "pad_token_id": 50256, | |
| "bos_token_id": 50257, | |
| "eos_token_id": 50257, | |
| "suppress_tokens": null, | |
| "begin_suppress_tokens": [ | |
| 220, | |
| 50257 | |
| ], | |
| "use_weighted_layer_sum": false, | |
| "classifier_proj_size": 256, | |
| "apply_spec_augment": false, | |
| "mask_time_prob": 0.05, | |
| "mask_time_length": 10, | |
| "mask_time_min_masks": 2, | |
| "mask_feature_prob": 0.0, | |
| "mask_feature_length": 10, | |
| "mask_feature_min_masks": 0, | |
| "median_filter_width": 7, | |
| "tie_word_embeddings": true, | |
| "_name_or_path": "openai/whisper-large-v3", | |
| "model_type": "whisper", | |
| "output_attentions": false | |
| }, | |
| "quant_config": { | |
| "components": [ | |
| "decoder" | |
| ], | |
| "scheme": "tritplane3", | |
| "group_size": 32, | |
| "n_iter": 10, | |
| "salient_fraction": 0.0, | |
| "rescue_fraction": 0.0, | |
| "n_planes": 3, | |
| "allow_all_linear": false, | |
| "target_module_names": [ | |
| "Wqkv", | |
| "att_proj", | |
| "attn.proj", | |
| "attn.qkv", | |
| "c_attn", | |
| "c_fc", | |
| "c_proj", | |
| "dense", | |
| "dense_4h_to_h", | |
| "dense_h_to_4h", | |
| "down_proj", | |
| "fc1", | |
| "fc2", | |
| "ff_proj", | |
| "gate_proj", | |
| "gate_up_proj", | |
| "k", | |
| "k_proj", | |
| "linear", | |
| "o", | |
| "o_proj", | |
| "out_proj", | |
| "per_layer_input_gate", | |
| "per_layer_projection", | |
| "proj", | |
| "q", | |
| "q_proj", | |
| "qkv", | |
| "qkv_proj", | |
| "query_key_value", | |
| "up_proj", | |
| "v", | |
| "v_proj", | |
| "w1", | |
| "w2", | |
| "w3", | |
| "wi", | |
| "wi_0", | |
| "wi_1", | |
| "wo" | |
| ], | |
| "max_length": 160, | |
| "calibration_batch_size": 2, | |
| "calibration_prompts": null, | |
| "vlm_use_demo_image": true | |
| }, | |
| "plan": {}, | |
| "layer_info": { | |
| "model.decoder.layers.0.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.0.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.1.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.2.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.3.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.4.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.5.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.6.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.7.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.8.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.9.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.10.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.11.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.12.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.13.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.14.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.15.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.16.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.17.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.18.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.19.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.20.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.21.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.22.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.23.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.24.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.25.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.26.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.27.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.28.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.29.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.30.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.self_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.out_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 1638400, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1843200, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.fc1": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 5120, | |
| 1280 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.decoder.layers.31.fc2": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1280, | |
| 5120 | |
| ], | |
| "dtype": "torch.float32", | |
| "num_elements": 6553600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7372800, | |
| "effective_bits": 9.0 | |
| } | |
| }, | |
| "stats": { | |
| "model.decoder.layers.0.self_attn.k_proj": { | |
| "mse": 1.8580032701720484e-05, | |
| "rmse": 0.004310456205753688, | |
| "relative_error": 0.09925386526034846, | |
| "max_error": 0.14437294006347656, | |
| "sparsity": 0.39224283893903095, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.self_attn.v_proj": { | |
| "mse": 2.4661758288857527e-06, | |
| "rmse": 0.001570406262368357, | |
| "relative_error": 0.11609086774916236, | |
| "max_error": 0.054907798767089844, | |
| "sparsity": 0.3929264346758525, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.self_attn.q_proj": { | |
| "mse": 1.6553260138607584e-05, | |
| "rmse": 0.0040685697903080865, | |
| "relative_error": 0.10149378917006845, | |
| "max_error": 0.1326584815979004, | |
| "sparsity": 0.3944903512795766, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.self_attn.out_proj": { | |
| "mse": 3.2385387385147624e-06, | |
| "rmse": 0.0017995940482549843, | |
| "relative_error": 0.13208308522140563, | |
| "max_error": 0.046822309494018555, | |
| "sparsity": 0.3820499777793884, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.k_proj": { | |
| "mse": 4.496247584029334e-06, | |
| "rmse": 0.00212043570617676, | |
| "relative_error": 0.13237802433595772, | |
| "max_error": 0.06512248516082764, | |
| "sparsity": 0.3777557412783305, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.v_proj": { | |
| "mse": 2.969502475025365e-06, | |
| "rmse": 0.0017232244412801732, | |
| "relative_error": 0.13877954132266926, | |
| "max_error": 0.10634016990661621, | |
| "sparsity": 0.3870005210240682, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.q_proj": { | |
| "mse": 4.127876309212297e-06, | |
| "rmse": 0.002031717576143962, | |
| "relative_error": 0.1009000601872321, | |
| "max_error": 0.06661701202392578, | |
| "sparsity": 0.3893473247687022, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.encoder_attn.out_proj": { | |
| "mse": 3.7137149320187746e-06, | |
| "rmse": 0.001927100135441533, | |
| "relative_error": 0.1675821035696163, | |
| "max_error": 0.07441520690917969, | |
| "sparsity": 0.36065856615702313, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.fc1": { | |
| "mse": 5.68257109989645e-06, | |
| "rmse": 0.002383814401310733, | |
| "relative_error": 0.1187574637829767, | |
| "max_error": 0.06409931182861328, | |
| "sparsity": 0.3814770579338074, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.0.fc2": { | |
| "mse": 1.4334982552099973e-05, | |
| "rmse": 0.0037861566993588596, | |
| "relative_error": 0.3031673577353539, | |
| "max_error": 0.16556644439697266, | |
| "sparsity": 0.3366134663422902, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.self_attn.k_proj": { | |
| "mse": 7.349524366873084e-06, | |
| "rmse": 0.0027110006209650865, | |
| "relative_error": 0.09842497300655098, | |
| "max_error": 0.06743669509887695, | |
| "sparsity": 0.383358766635259, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.self_attn.v_proj": { | |
| "mse": 1.190571083498071e-06, | |
| "rmse": 0.0010911329357590077, | |
| "relative_error": 0.11816144576956879, | |
| "max_error": 0.0364990234375, | |
| "sparsity": 0.4012465476989746, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.self_attn.q_proj": { | |
| "mse": 9.280999620386865e-06, | |
| "rmse": 0.0030464733086614866, | |
| "relative_error": 0.12041031868367, | |
| "max_error": 0.09621429443359375, | |
| "sparsity": 0.39007466038068134, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.self_attn.out_proj": { | |
| "mse": 2.9960222036606865e-06, | |
| "rmse": 0.0017309021357837324, | |
| "relative_error": 0.18230374975284447, | |
| "max_error": 0.06497567892074585, | |
| "sparsity": 0.36671244104703266, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.k_proj": { | |
| "mse": 4.528291810856899e-06, | |
| "rmse": 0.0021279783389068834, | |
| "relative_error": 0.13069245935407484, | |
| "max_error": 0.07462644577026367, | |
| "sparsity": 0.3784716725349426, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.v_proj": { | |
| "mse": 3.1088975447346456e-06, | |
| "rmse": 0.001763206608635144, | |
| "relative_error": 0.13963361561705537, | |
| "max_error": 0.045983314514160156, | |
| "sparsity": 0.3875199357668559, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.q_proj": { | |
| "mse": 3.4988427159987623e-06, | |
| "rmse": 0.0018705193706558515, | |
| "relative_error": 0.0942609671296887, | |
| "max_error": 0.05310773849487305, | |
| "sparsity": 0.3917895456155141, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.encoder_attn.out_proj": { | |
| "mse": 3.4424929253873415e-06, | |
| "rmse": 0.0018553956250318532, | |
| "relative_error": 0.15703971142469872, | |
| "max_error": 0.05678129196166992, | |
| "sparsity": 0.3643878102302551, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.fc1": { | |
| "mse": 6.53987035548198e-06, | |
| "rmse": 0.0025573170228741647, | |
| "relative_error": 0.1336126573774987, | |
| "max_error": 0.12996280193328857, | |
| "sparsity": 0.3849823971589406, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.1.fc2": { | |
| "mse": 9.913981557474472e-06, | |
| "rmse": 0.0031486475759402595, | |
| "relative_error": 0.26228266810436957, | |
| "max_error": 0.16301476955413818, | |
| "sparsity": 0.36005385716756183, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.self_attn.k_proj": { | |
| "mse": 6.23207597527653e-06, | |
| "rmse": 0.002496412621197972, | |
| "relative_error": 0.09934420685860854, | |
| "max_error": 0.089263916015625, | |
| "sparsity": 0.3799009323120117, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.self_attn.v_proj": { | |
| "mse": 2.02669070858974e-06, | |
| "rmse": 0.0014236188775756454, | |
| "relative_error": 0.14201269459831478, | |
| "max_error": 0.05799412727355957, | |
| "sparsity": 0.39388997356096905, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.self_attn.q_proj": { | |
| "mse": 6.305576334852958e-06, | |
| "rmse": 0.0025110906663943773, | |
| "relative_error": 0.10300231619558352, | |
| "max_error": 0.0862736701965332, | |
| "sparsity": 0.39087076981862384, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.self_attn.out_proj": { | |
| "mse": 4.497860118135577e-06, | |
| "rmse": 0.0021208159085916857, | |
| "relative_error": 0.2195022799922506, | |
| "max_error": 0.3148527145385742, | |
| "sparsity": 0.3396537204583486, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.k_proj": { | |
| "mse": 2.4377934551012004e-06, | |
| "rmse": 0.001561343477618298, | |
| "relative_error": 0.12683146014052707, | |
| "max_error": 0.059914588928222656, | |
| "sparsity": 0.37750569979349774, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.v_proj": { | |
| "mse": 2.0136199054832105e-06, | |
| "rmse": 0.0014190207558324192, | |
| "relative_error": 0.13852380424527688, | |
| "max_error": 0.05537229776382446, | |
| "sparsity": 0.3873927891254425, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.q_proj": { | |
| "mse": 2.6865686777455267e-06, | |
| "rmse": 0.0016390755558379627, | |
| "relative_error": 0.10038211344163599, | |
| "max_error": 0.07483410835266113, | |
| "sparsity": 0.38270671168963116, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.encoder_attn.out_proj": { | |
| "mse": 1.6374598317270284e-06, | |
| "rmse": 0.0012796326940677269, | |
| "relative_error": 0.14195360839101503, | |
| "max_error": 0.04471588134765625, | |
| "sparsity": 0.36931029955546063, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.fc1": { | |
| "mse": 8.688674824952614e-06, | |
| "rmse": 0.0029476558186044403, | |
| "relative_error": 0.14342776469850538, | |
| "max_error": 0.23823362588882446, | |
| "sparsity": 0.38876607020696, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.2.fc2": { | |
| "mse": 1.0341858796891756e-05, | |
| "rmse": 0.003215876054342231, | |
| "relative_error": 0.24973688000903327, | |
| "max_error": 0.3158590793609619, | |
| "sparsity": 0.3736262520154317, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.self_attn.k_proj": { | |
| "mse": 4.288419859221904e-06, | |
| "rmse": 0.0020708500330110587, | |
| "relative_error": 0.1155409851818616, | |
| "max_error": 0.06696194410324097, | |
| "sparsity": 0.3892299433549245, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.self_attn.v_proj": { | |
| "mse": 2.573929577920353e-06, | |
| "rmse": 0.0016043470877339333, | |
| "relative_error": 0.19526253426874657, | |
| "max_error": 0.06625974178314209, | |
| "sparsity": 0.4321826100349426, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.self_attn.q_proj": { | |
| "mse": 4.509508016781183e-06, | |
| "rmse": 0.0021235602220754613, | |
| "relative_error": 0.09801006840201408, | |
| "max_error": 0.1561570167541504, | |
| "sparsity": 0.4001656075318654, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.self_attn.out_proj": { | |
| "mse": 3.5177963582100347e-06, | |
| "rmse": 0.0018755789394770978, | |
| "relative_error": 0.24069881019706324, | |
| "max_error": 0.08281075954437256, | |
| "sparsity": 0.3444596230983734, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.k_proj": { | |
| "mse": 1.4662796274933498e-06, | |
| "rmse": 0.0012109003375560476, | |
| "relative_error": 0.12159416050147576, | |
| "max_error": 0.07092952728271484, | |
| "sparsity": 0.3794936041037242, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.v_proj": { | |
| "mse": 1.4209160781319952e-06, | |
| "rmse": 0.0011920218446538618, | |
| "relative_error": 0.1381781999556332, | |
| "max_error": 0.04003572463989258, | |
| "sparsity": 0.386987308661143, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.q_proj": { | |
| "mse": 1.2005122016489622e-06, | |
| "rmse": 0.0010956788770661605, | |
| "relative_error": 0.09110538562732882, | |
| "max_error": 0.04202413558959961, | |
| "sparsity": 0.4271486500898997, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.encoder_attn.out_proj": { | |
| "mse": 9.809649554881616e-07, | |
| "rmse": 0.0009904367498675328, | |
| "relative_error": 0.137609858502114, | |
| "max_error": 0.0486452579498291, | |
| "sparsity": 0.3676226834456126, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.fc1": { | |
| "mse": 8.359793355339207e-06, | |
| "rmse": 0.002891330723964176, | |
| "relative_error": 0.13638459317304816, | |
| "max_error": 0.143798828125, | |
| "sparsity": 0.3915335138638814, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.3.fc2": { | |
| "mse": 1.5966204955475405e-05, | |
| "rmse": 0.0039957733864016114, | |
| "relative_error": 0.2976195826777279, | |
| "max_error": 0.23508834838867188, | |
| "sparsity": 0.33907978733380634, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.self_attn.k_proj": { | |
| "mse": 5.826852429891005e-06, | |
| "rmse": 0.0024138874103592747, | |
| "relative_error": 0.13315749710257516, | |
| "max_error": 0.09303951263427734, | |
| "sparsity": 0.3802773058414459, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.self_attn.v_proj": { | |
| "mse": 4.074689513799967e-06, | |
| "rmse": 0.0020185860184297245, | |
| "relative_error": 0.17622768053377313, | |
| "max_error": 0.06643867492675781, | |
| "sparsity": 0.4024471044540405, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.self_attn.q_proj": { | |
| "mse": 6.895251772220945e-06, | |
| "rmse": 0.0026258811420589744, | |
| "relative_error": 0.11312335190487666, | |
| "max_error": 0.095916748046875, | |
| "sparsity": 0.39481709400812787, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.self_attn.out_proj": { | |
| "mse": 3.94699236494489e-06, | |
| "rmse": 0.0019867038946317314, | |
| "relative_error": 0.19971364948092138, | |
| "max_error": 0.14113712310791016, | |
| "sparsity": 0.36342406272888184, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.k_proj": { | |
| "mse": 3.741293539860635e-06, | |
| "rmse": 0.0019342423684379979, | |
| "relative_error": 0.13165244864760506, | |
| "max_error": 0.10697746276855469, | |
| "sparsity": 0.3745935062567393, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.v_proj": { | |
| "mse": 2.3680458980379626e-06, | |
| "rmse": 0.0015388456381450228, | |
| "relative_error": 0.13682706545378434, | |
| "max_error": 0.06878548860549927, | |
| "sparsity": 0.38733479380607605, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.q_proj": { | |
| "mse": 3.984724571637344e-06, | |
| "rmse": 0.001996177490013687, | |
| "relative_error": 0.10924744621771405, | |
| "max_error": 0.12373250722885132, | |
| "sparsity": 0.3910862108071645, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.encoder_attn.out_proj": { | |
| "mse": 2.085837877530139e-06, | |
| "rmse": 0.0014442430119374437, | |
| "relative_error": 0.1408655786530582, | |
| "max_error": 0.054406166076660156, | |
| "sparsity": 0.371499627828598, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.fc1": { | |
| "mse": 9.162421520159114e-06, | |
| "rmse": 0.0030269492100395597, | |
| "relative_error": 0.1464674758594165, | |
| "max_error": 0.18100547790527344, | |
| "sparsity": 0.3841899633407593, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.4.fc2": { | |
| "mse": 8.689513379067648e-06, | |
| "rmse": 0.0029477980560187037, | |
| "relative_error": 0.26735733458475613, | |
| "max_error": 0.26274871826171875, | |
| "sparsity": 0.3929274578889211, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.self_attn.k_proj": { | |
| "mse": 7.255655873450451e-06, | |
| "rmse": 0.002693632468146026, | |
| "relative_error": 0.11698810685153418, | |
| "max_error": 0.0920257568359375, | |
| "sparsity": 0.3831644654273987, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.self_attn.v_proj": { | |
| "mse": 3.6478245419857558e-06, | |
| "rmse": 0.0019099278892109398, | |
| "relative_error": 0.1353195932371794, | |
| "max_error": 0.06800031661987305, | |
| "sparsity": 0.39072123169898987, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.self_attn.q_proj": { | |
| "mse": 8.386698027607054e-06, | |
| "rmse": 0.0028959796317666073, | |
| "relative_error": 0.11046416128292798, | |
| "max_error": 0.13091278076171875, | |
| "sparsity": 0.38711018363634747, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.self_attn.out_proj": { | |
| "mse": 2.741804564720951e-06, | |
| "rmse": 0.0016558395347137207, | |
| "relative_error": 0.12715778956408427, | |
| "max_error": 0.04480600357055664, | |
| "sparsity": 0.37403035163879395, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.k_proj": { | |
| "mse": 2.9920131510152714e-06, | |
| "rmse": 0.001729743666274073, | |
| "relative_error": 0.12890427572191027, | |
| "max_error": 0.07967853546142578, | |
| "sparsity": 0.376779168844223, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.v_proj": { | |
| "mse": 2.630024482641602e-06, | |
| "rmse": 0.001621735022326891, | |
| "relative_error": 0.13842549036657706, | |
| "max_error": 0.05739307403564453, | |
| "sparsity": 0.38748615980148315, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.q_proj": { | |
| "mse": 3.7650345348083647e-06, | |
| "rmse": 0.0019403696902416212, | |
| "relative_error": 0.1053177541776783, | |
| "max_error": 0.070953369140625, | |
| "sparsity": 0.3865022857983907, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.encoder_attn.out_proj": { | |
| "mse": 2.2414869818021543e-06, | |
| "rmse": 0.0014971596380487133, | |
| "relative_error": 0.1441333736775373, | |
| "max_error": 0.04926919937133789, | |
| "sparsity": 0.3707836965719859, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.fc1": { | |
| "mse": 6.184009635035181e-06, | |
| "rmse": 0.002486766904041306, | |
| "relative_error": 0.12856168122610537, | |
| "max_error": 0.10060501098632812, | |
| "sparsity": 0.3779651920000712, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.5.fc2": { | |
| "mse": 7.644514880666975e-06, | |
| "rmse": 0.0027648715848420475, | |
| "relative_error": 0.23591981790548097, | |
| "max_error": 0.23477983474731445, | |
| "sparsity": 0.39138280351956684, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.self_attn.k_proj": { | |
| "mse": 7.090200142556569e-06, | |
| "rmse": 0.0026627429734310763, | |
| "relative_error": 0.10550321779770969, | |
| "max_error": 0.085968017578125, | |
| "sparsity": 0.3890014588832855, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.self_attn.v_proj": { | |
| "mse": 2.71016028818849e-06, | |
| "rmse": 0.0016462564466657344, | |
| "relative_error": 0.11301504862523593, | |
| "max_error": 0.04952096939086914, | |
| "sparsity": 0.39544881383577984, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.self_attn.q_proj": { | |
| "mse": 7.975068911036942e-06, | |
| "rmse": 0.0028240164502065035, | |
| "relative_error": 0.10586660338207213, | |
| "max_error": 0.10039472579956055, | |
| "sparsity": 0.392013152440389, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.self_attn.out_proj": { | |
| "mse": 3.0777111987845274e-06, | |
| "rmse": 0.0017543406735251073, | |
| "relative_error": 0.12534030843378727, | |
| "max_error": 0.0577239990234375, | |
| "sparsity": 0.37753153840700787, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.k_proj": { | |
| "mse": 4.493654159887228e-06, | |
| "rmse": 0.002119824087014587, | |
| "relative_error": 0.13203263972544632, | |
| "max_error": 0.091705322265625, | |
| "sparsity": 0.3797082503636678, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.v_proj": { | |
| "mse": 3.07243135466706e-06, | |
| "rmse": 0.0017528352331771118, | |
| "relative_error": 0.1375799140693424, | |
| "max_error": 0.06963682174682617, | |
| "sparsity": 0.38788431882858276, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.q_proj": { | |
| "mse": 3.828226908808574e-06, | |
| "rmse": 0.001956585522998822, | |
| "relative_error": 0.10295843016429852, | |
| "max_error": 0.06548118591308594, | |
| "sparsity": 0.39143106341362, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.encoder_attn.out_proj": { | |
| "mse": 3.1550871426588856e-06, | |
| "rmse": 0.0017762564968660595, | |
| "relative_error": 0.1498996373199617, | |
| "max_error": 0.06113243103027344, | |
| "sparsity": 0.3675134281317393, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.fc1": { | |
| "mse": 6.103840405558003e-06, | |
| "rmse": 0.002470595152095544, | |
| "relative_error": 0.12313088356676909, | |
| "max_error": 0.1277751922607422, | |
| "sparsity": 0.3875523904959361, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.6.fc2": { | |
| "mse": 5.892592071177205e-06, | |
| "rmse": 0.0024274661833230972, | |
| "relative_error": 0.19870620551361595, | |
| "max_error": 0.12247085571289062, | |
| "sparsity": 0.3763323525587718, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.self_attn.k_proj": { | |
| "mse": 7.188941708591301e-06, | |
| "rmse": 0.002681220190247586, | |
| "relative_error": 0.10391021716845518, | |
| "max_error": 0.09998416900634766, | |
| "sparsity": 0.38841044902801514, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.self_attn.v_proj": { | |
| "mse": 2.650939677550923e-06, | |
| "rmse": 0.0016281706536941769, | |
| "relative_error": 0.11203954726599678, | |
| "max_error": 0.04358863830566406, | |
| "sparsity": 0.39471496144930523, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.self_attn.q_proj": { | |
| "mse": 7.50063918530941e-06, | |
| "rmse": 0.002738729483776996, | |
| "relative_error": 0.10596623949152305, | |
| "max_error": 0.08472824096679688, | |
| "sparsity": 0.3926519850889842, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.self_attn.out_proj": { | |
| "mse": 3.575294613256119e-06, | |
| "rmse": 0.0018908449469102746, | |
| "relative_error": 0.13632958989726185, | |
| "max_error": 0.15756988525390625, | |
| "sparsity": 0.3729923566182454, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.k_proj": { | |
| "mse": 5.2371242418303154e-06, | |
| "rmse": 0.0022884764018513094, | |
| "relative_error": 0.1284623026285162, | |
| "max_error": 0.08586883544921875, | |
| "sparsity": 0.37997640172640484, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.v_proj": { | |
| "mse": 3.908655344275758e-06, | |
| "rmse": 0.0019770319532763647, | |
| "relative_error": 0.1394144982713944, | |
| "max_error": 0.051363348960876465, | |
| "sparsity": 0.38741718729337055, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.q_proj": { | |
| "mse": 4.317747880122624e-06, | |
| "rmse": 0.002077919122613444, | |
| "relative_error": 0.10485210988714737, | |
| "max_error": 0.07918024063110352, | |
| "sparsity": 0.38702677687009174, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.encoder_attn.out_proj": { | |
| "mse": 3.5595121516962536e-06, | |
| "rmse": 0.0018866669424400942, | |
| "relative_error": 0.14335214025840035, | |
| "max_error": 0.049762725830078125, | |
| "sparsity": 0.370357871055603, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.fc1": { | |
| "mse": 5.996387244522339e-06, | |
| "rmse": 0.0024487521811164025, | |
| "relative_error": 0.1205863772801864, | |
| "max_error": 0.11818951368331909, | |
| "sparsity": 0.38447315494219464, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.7.fc2": { | |
| "mse": 6.120802936493419e-06, | |
| "rmse": 0.002474025653968329, | |
| "relative_error": 0.1946466951576888, | |
| "max_error": 0.12637996673583984, | |
| "sparsity": 0.3830512960751851, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.self_attn.k_proj": { | |
| "mse": 9.389744263899047e-06, | |
| "rmse": 0.0030642689607635696, | |
| "relative_error": 0.10362220973376526, | |
| "max_error": 0.15833282470703125, | |
| "sparsity": 0.38879414399464923, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.self_attn.v_proj": { | |
| "mse": 3.204965423719841e-06, | |
| "rmse": 0.001790241722148113, | |
| "relative_error": 0.11599168622346309, | |
| "max_error": 0.04339766502380371, | |
| "sparsity": 0.39585408568382263, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.self_attn.q_proj": { | |
| "mse": 9.705722732178401e-06, | |
| "rmse": 0.0031154008942956926, | |
| "relative_error": 0.1066208366419977, | |
| "max_error": 0.15021896362304688, | |
| "sparsity": 0.39206990599632263, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.self_attn.out_proj": { | |
| "mse": 3.765134124478209e-06, | |
| "rmse": 0.0019403953526222972, | |
| "relative_error": 0.1315258084028414, | |
| "max_error": 0.07892656326293945, | |
| "sparsity": 0.3757346570491791, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.k_proj": { | |
| "mse": 3.4628087632881943e-06, | |
| "rmse": 0.0018608623708614763, | |
| "relative_error": 0.12595009535512863, | |
| "max_error": 0.09252160787582397, | |
| "sparsity": 0.37416035930315655, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.v_proj": { | |
| "mse": 2.5030187771335477e-06, | |
| "rmse": 0.001582093163228243, | |
| "relative_error": 0.13727063287058378, | |
| "max_error": 0.10220539569854736, | |
| "sparsity": 0.3874037762482961, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.q_proj": { | |
| "mse": 3.3098947369580856e-06, | |
| "rmse": 0.0018193116107357984, | |
| "relative_error": 0.1058808222857755, | |
| "max_error": 0.06249856948852539, | |
| "sparsity": 0.3892553746700287, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.encoder_attn.out_proj": { | |
| "mse": 2.3032778244669316e-06, | |
| "rmse": 0.0015176553707831472, | |
| "relative_error": 0.15423321115473757, | |
| "max_error": 0.04558539390563965, | |
| "sparsity": 0.3738443950812022, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.fc1": { | |
| "mse": 6.487789960374357e-06, | |
| "rmse": 0.0025471140454197093, | |
| "relative_error": 0.1214262870866511, | |
| "max_error": 0.15251731872558594, | |
| "sparsity": 0.3895666499932607, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.8.fc2": { | |
| "mse": 4.937920493830461e-06, | |
| "rmse": 0.00222214322081869, | |
| "relative_error": 0.1673341818290483, | |
| "max_error": 0.10040855407714844, | |
| "sparsity": 0.38012537360191345, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.self_attn.k_proj": { | |
| "mse": 7.597804142278619e-06, | |
| "rmse": 0.0027564114609902888, | |
| "relative_error": 0.10317176818694523, | |
| "max_error": 0.10155200958251953, | |
| "sparsity": 0.38914082447687787, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.self_attn.v_proj": { | |
| "mse": 3.0955325200920925e-06, | |
| "rmse": 0.0017594125497142768, | |
| "relative_error": 0.11151948536587492, | |
| "max_error": 0.04322338104248047, | |
| "sparsity": 0.39580180247624713, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.self_attn.q_proj": { | |
| "mse": 8.065996553341392e-06, | |
| "rmse": 0.0028400698148709994, | |
| "relative_error": 0.10295286088972519, | |
| "max_error": 0.0933065414428711, | |
| "sparsity": 0.39085469643274945, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.self_attn.out_proj": { | |
| "mse": 3.508735062496271e-06, | |
| "rmse": 0.0018731617822538103, | |
| "relative_error": 0.12311608977593916, | |
| "max_error": 0.04478812217712402, | |
| "sparsity": 0.3779345651467641, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.k_proj": { | |
| "mse": 4.8464730753039476e-06, | |
| "rmse": 0.0022014706619221495, | |
| "relative_error": 0.13032986898921775, | |
| "max_error": 0.09057807922363281, | |
| "sparsity": 0.37835489710172016, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.v_proj": { | |
| "mse": 3.663315510493703e-06, | |
| "rmse": 0.0019139789733677074, | |
| "relative_error": 0.13948631905897965, | |
| "max_error": 0.07102203369140625, | |
| "sparsity": 0.3872808913389842, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.q_proj": { | |
| "mse": 4.2114693314943e-06, | |
| "rmse": 0.002052186475809228, | |
| "relative_error": 0.10250248242039318, | |
| "max_error": 0.07402801513671875, | |
| "sparsity": 0.3858288526535034, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.encoder_attn.out_proj": { | |
| "mse": 3.6693884339911165e-06, | |
| "rmse": 0.0019155647819875778, | |
| "relative_error": 0.1500559429813565, | |
| "max_error": 0.04814910888671875, | |
| "sparsity": 0.3671809832255046, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.fc1": { | |
| "mse": 6.263318937271833e-06, | |
| "rmse": 0.002502662369811764, | |
| "relative_error": 0.11967164179023133, | |
| "max_error": 0.13221168518066406, | |
| "sparsity": 0.3938055435816447, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.9.fc2": { | |
| "mse": 5.56699978915276e-06, | |
| "rmse": 0.002359449043559271, | |
| "relative_error": 0.167287240984828, | |
| "max_error": 0.16849833726882935, | |
| "sparsity": 0.37245218952496845, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.self_attn.k_proj": { | |
| "mse": 7.05444017512491e-06, | |
| "rmse": 0.0026560196112086426, | |
| "relative_error": 0.10128810493124905, | |
| "max_error": 0.08446979522705078, | |
| "sparsity": 0.39093485474586487, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.self_attn.v_proj": { | |
| "mse": 2.5130254925898043e-06, | |
| "rmse": 0.0015852525012091305, | |
| "relative_error": 0.102845789830796, | |
| "max_error": 0.055767059326171875, | |
| "sparsity": 0.395834356546402, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.self_attn.q_proj": { | |
| "mse": 7.904863196017686e-06, | |
| "rmse": 0.0028115588551580572, | |
| "relative_error": 0.10666928453972531, | |
| "max_error": 0.10036659240722656, | |
| "sparsity": 0.39256551861763, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.self_attn.out_proj": { | |
| "mse": 3.6509111396298977e-06, | |
| "rmse": 0.001910735758714401, | |
| "relative_error": 0.12703838302529682, | |
| "max_error": 0.05091357231140137, | |
| "sparsity": 0.37751484910647076, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.k_proj": { | |
| "mse": 5.80274945605197e-06, | |
| "rmse": 0.002408889672868388, | |
| "relative_error": 0.13063022301510013, | |
| "max_error": 0.15572518110275269, | |
| "sparsity": 0.3812152047952016, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.v_proj": { | |
| "mse": 3.620761845013476e-06, | |
| "rmse": 0.0019028299569361093, | |
| "relative_error": 0.1386701153046746, | |
| "max_error": 0.11056578159332275, | |
| "sparsity": 0.38714579741160077, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.q_proj": { | |
| "mse": 4.344560238678241e-06, | |
| "rmse": 0.002084360870549589, | |
| "relative_error": 0.106447564430133, | |
| "max_error": 0.09343606233596802, | |
| "sparsity": 0.390997310479482, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.encoder_attn.out_proj": { | |
| "mse": 3.1880676942819264e-06, | |
| "rmse": 0.0017855160862568352, | |
| "relative_error": 0.13966577394841476, | |
| "max_error": 0.0576634407043457, | |
| "sparsity": 0.3778953055540721, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.fc1": { | |
| "mse": 5.8392433857079595e-06, | |
| "rmse": 0.0024164526450373407, | |
| "relative_error": 0.11521382568878719, | |
| "max_error": 0.1955888867378235, | |
| "sparsity": 0.3927546739578247, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.10.fc2": { | |
| "mse": 5.1562446969910525e-06, | |
| "rmse": 0.002270736597888679, | |
| "relative_error": 0.16286670496747932, | |
| "max_error": 0.10766124725341797, | |
| "sparsity": 0.37760167320569354, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.self_attn.k_proj": { | |
| "mse": 7.5763182394439355e-06, | |
| "rmse": 0.002752511260548072, | |
| "relative_error": 0.10028402369879523, | |
| "max_error": 0.11373233795166016, | |
| "sparsity": 0.3905513485272725, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.self_attn.v_proj": { | |
| "mse": 2.6552777399047045e-06, | |
| "rmse": 0.0016295022982201358, | |
| "relative_error": 0.10318013366971909, | |
| "max_error": 0.03563690185546875, | |
| "sparsity": 0.396579384803772, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.self_attn.q_proj": { | |
| "mse": 8.213575711124577e-06, | |
| "rmse": 0.0028659336543480165, | |
| "relative_error": 0.10528111671620048, | |
| "max_error": 0.13054275512695312, | |
| "sparsity": 0.3914807140827179, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.self_attn.out_proj": { | |
| "mse": 3.830523837677902e-06, | |
| "rmse": 0.001957172408776984, | |
| "relative_error": 0.12762631289255238, | |
| "max_error": 0.06855440139770508, | |
| "sparsity": 0.37757569551467896, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.k_proj": { | |
| "mse": 6.9311131483118515e-06, | |
| "rmse": 0.002632700732766991, | |
| "relative_error": 0.1296334028364033, | |
| "max_error": 0.0937509536743164, | |
| "sparsity": 0.38014403978983563, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.v_proj": { | |
| "mse": 4.078871370438719e-06, | |
| "rmse": 0.0020196215909022955, | |
| "relative_error": 0.13855830017767207, | |
| "max_error": 0.05998337268829346, | |
| "sparsity": 0.3870062232017517, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.q_proj": { | |
| "mse": 4.832721970160492e-06, | |
| "rmse": 0.00219834528001415, | |
| "relative_error": 0.10118615123234728, | |
| "max_error": 0.07009029388427734, | |
| "sparsity": 0.39061442017555237, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.encoder_attn.out_proj": { | |
| "mse": 4.546827767626382e-06, | |
| "rmse": 0.0021323291883821273, | |
| "relative_error": 0.1541347341595871, | |
| "max_error": 0.08632850646972656, | |
| "sparsity": 0.36755330363909405, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.fc1": { | |
| "mse": 5.669960046361666e-06, | |
| "rmse": 0.002381167790467876, | |
| "relative_error": 0.11277456767040349, | |
| "max_error": 0.18575668334960938, | |
| "sparsity": 0.39261088768641156, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.11.fc2": { | |
| "mse": 5.671568032994401e-06, | |
| "rmse": 0.0023815054131776396, | |
| "relative_error": 0.16865825905003715, | |
| "max_error": 0.15157604217529297, | |
| "sparsity": 0.3780089815457662, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.self_attn.k_proj": { | |
| "mse": 7.2157231443270575e-06, | |
| "rmse": 0.002686209810183683, | |
| "relative_error": 0.10297321982945866, | |
| "max_error": 0.14776229858398438, | |
| "sparsity": 0.3929532965024312, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.self_attn.v_proj": { | |
| "mse": 2.2166418602864724e-06, | |
| "rmse": 0.0014888390981857215, | |
| "relative_error": 0.10166004020558055, | |
| "max_error": 0.0788717269897461, | |
| "sparsity": 0.39558594425519306, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.self_attn.q_proj": { | |
| "mse": 7.965501936268993e-06, | |
| "rmse": 0.0028223220823054536, | |
| "relative_error": 0.10574771075147456, | |
| "max_error": 0.12176895141601562, | |
| "sparsity": 0.39267314473787945, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.self_attn.out_proj": { | |
| "mse": 4.219202310196124e-06, | |
| "rmse": 0.0020540696945810103, | |
| "relative_error": 0.14246199970455647, | |
| "max_error": 0.04717373847961426, | |
| "sparsity": 0.37325459718704224, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.k_proj": { | |
| "mse": 7.433164682879578e-06, | |
| "rmse": 0.0027263830770600777, | |
| "relative_error": 0.12968853035910288, | |
| "max_error": 0.09462738037109375, | |
| "sparsity": 0.38084818919499713, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.v_proj": { | |
| "mse": 4.1861885620164685e-06, | |
| "rmse": 0.002046017732576252, | |
| "relative_error": 0.13816164773288764, | |
| "max_error": 0.05682635307312012, | |
| "sparsity": 0.38716431458791095, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.q_proj": { | |
| "mse": 4.936920049658511e-06, | |
| "rmse": 0.0022219181014741544, | |
| "relative_error": 0.10136127506814356, | |
| "max_error": 0.0756378173828125, | |
| "sparsity": 0.3903729220231374, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.encoder_attn.out_proj": { | |
| "mse": 4.398137662064983e-06, | |
| "rmse": 0.002097173731970001, | |
| "relative_error": 0.15215107280823817, | |
| "max_error": 0.04639291763305664, | |
| "sparsity": 0.3681211272875468, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.fc1": { | |
| "mse": 5.270377641863888e-06, | |
| "rmse": 0.0022957303068661806, | |
| "relative_error": 0.11048733198883895, | |
| "max_error": 0.19587993621826172, | |
| "sparsity": 0.3938109278678894, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.12.fc2": { | |
| "mse": 4.693832579505397e-06, | |
| "rmse": 0.0021665254624641263, | |
| "relative_error": 0.15548740257607477, | |
| "max_error": 0.1594524383544922, | |
| "sparsity": 0.3831084171930949, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.self_attn.k_proj": { | |
| "mse": 7.437230124196503e-06, | |
| "rmse": 0.0027271285492613843, | |
| "relative_error": 0.10205442810215265, | |
| "max_error": 0.10116291046142578, | |
| "sparsity": 0.3928145269552867, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.self_attn.v_proj": { | |
| "mse": 2.572279754531337e-06, | |
| "rmse": 0.0016038328324770438, | |
| "relative_error": 0.10003899297347467, | |
| "max_error": 0.046753883361816406, | |
| "sparsity": 0.3959210216999054, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.self_attn.q_proj": { | |
| "mse": 8.301682100864127e-06, | |
| "rmse": 0.002881263976254888, | |
| "relative_error": 0.10354114439325438, | |
| "max_error": 0.08680534362792969, | |
| "sparsity": 0.39322590827941895, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.self_attn.out_proj": { | |
| "mse": 4.205505319987424e-06, | |
| "rmse": 0.0020507328738739775, | |
| "relative_error": 0.13017285139173607, | |
| "max_error": 0.0489422082901001, | |
| "sparsity": 0.3762286305427551, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.k_proj": { | |
| "mse": 8.00198995420942e-06, | |
| "rmse": 0.002828778880402182, | |
| "relative_error": 0.1302027126358799, | |
| "max_error": 0.09444046020507812, | |
| "sparsity": 0.38099955519040424, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.v_proj": { | |
| "mse": 4.917752448818646e-06, | |
| "rmse": 0.0022176006062451024, | |
| "relative_error": 0.14032503322385256, | |
| "max_error": 0.1527886986732483, | |
| "sparsity": 0.38753193616867065, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.q_proj": { | |
| "mse": 4.973252544004936e-06, | |
| "rmse": 0.002230079044340118, | |
| "relative_error": 0.10241157821155644, | |
| "max_error": 0.07323455810546875, | |
| "sparsity": 0.38992024461428326, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.encoder_attn.out_proj": { | |
| "mse": 4.494896074902499e-06, | |
| "rmse": 0.002120116995569466, | |
| "relative_error": 0.14295791177064576, | |
| "max_error": 0.059726715087890625, | |
| "sparsity": 0.36936360597610474, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.fc1": { | |
| "mse": 4.528509634837974e-06, | |
| "rmse": 0.0021280295192590663, | |
| "relative_error": 0.10498195147758588, | |
| "max_error": 0.10173201560974121, | |
| "sparsity": 0.39298691352208454, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.13.fc2": { | |
| "mse": 3.464268047537189e-06, | |
| "rmse": 0.001861254428480209, | |
| "relative_error": 0.1368039681605119, | |
| "max_error": 0.12881851196289062, | |
| "sparsity": 0.3906531234582265, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.self_attn.k_proj": { | |
| "mse": 7.007335625530686e-06, | |
| "rmse": 0.0026471372509808944, | |
| "relative_error": 0.10088400764888705, | |
| "max_error": 0.09189605712890625, | |
| "sparsity": 0.3921105961004893, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.self_attn.v_proj": { | |
| "mse": 2.663620762177743e-06, | |
| "rmse": 0.0016320602814166342, | |
| "relative_error": 0.1011923521633487, | |
| "max_error": 0.03562361001968384, | |
| "sparsity": 0.3958081106344859, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.self_attn.q_proj": { | |
| "mse": 7.616056336701149e-06, | |
| "rmse": 0.0027597203366828945, | |
| "relative_error": 0.10217126831633329, | |
| "max_error": 0.10458946228027344, | |
| "sparsity": 0.3935355643431346, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.self_attn.out_proj": { | |
| "mse": 4.26402039011009e-06, | |
| "rmse": 0.0020649504570594644, | |
| "relative_error": 0.1307984487758854, | |
| "max_error": 0.06392478942871094, | |
| "sparsity": 0.3763623038927714, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.k_proj": { | |
| "mse": 8.15799012343632e-06, | |
| "rmse": 0.0028562195509862893, | |
| "relative_error": 0.13127130030753512, | |
| "max_error": 0.10230636596679688, | |
| "sparsity": 0.38005390763282776, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.v_proj": { | |
| "mse": 4.711388100986369e-06, | |
| "rmse": 0.0021705732194483487, | |
| "relative_error": 0.14000454719779204, | |
| "max_error": 0.1014096736907959, | |
| "sparsity": 0.38785481452941895, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.q_proj": { | |
| "mse": 5.099094778415747e-06, | |
| "rmse": 0.002258117529805689, | |
| "relative_error": 0.10179353870182929, | |
| "max_error": 0.0604557991027832, | |
| "sparsity": 0.39034892121950787, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.encoder_attn.out_proj": { | |
| "mse": 4.205314780847402e-06, | |
| "rmse": 0.002050686416994905, | |
| "relative_error": 0.14174980080853855, | |
| "max_error": 0.051389217376708984, | |
| "sparsity": 0.3708428939183553, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.fc1": { | |
| "mse": 4.525718850345584e-06, | |
| "rmse": 0.0021273736978597775, | |
| "relative_error": 0.10567714288451298, | |
| "max_error": 0.0731668472290039, | |
| "sparsity": 0.3928904136021932, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.14.fc2": { | |
| "mse": 3.092567112616962e-06, | |
| "rmse": 0.001758569621202687, | |
| "relative_error": 0.12642919819468282, | |
| "max_error": 0.1142416000366211, | |
| "sparsity": 0.38989150524139404, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.self_attn.k_proj": { | |
| "mse": 7.014393304416444e-06, | |
| "rmse": 0.0026484699931123338, | |
| "relative_error": 0.10064809834419954, | |
| "max_error": 0.1264019012451172, | |
| "sparsity": 0.3922804792722066, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.self_attn.v_proj": { | |
| "mse": 2.6377790618425934e-06, | |
| "rmse": 0.0016241240906539726, | |
| "relative_error": 0.10085383768859026, | |
| "max_error": 0.07508617639541626, | |
| "sparsity": 0.3954125940799713, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.self_attn.q_proj": { | |
| "mse": 7.531468781962758e-06, | |
| "rmse": 0.0027443521607043724, | |
| "relative_error": 0.10419899515710726, | |
| "max_error": 0.13496017456054688, | |
| "sparsity": 0.39298726121584576, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.self_attn.out_proj": { | |
| "mse": 4.675791842601029e-06, | |
| "rmse": 0.0021623579358193753, | |
| "relative_error": 0.1357531796701401, | |
| "max_error": 0.07401752471923828, | |
| "sparsity": 0.3740566074848175, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.k_proj": { | |
| "mse": 8.831677405396476e-06, | |
| "rmse": 0.002971813824147885, | |
| "relative_error": 0.1312616506673937, | |
| "max_error": 0.10747861862182617, | |
| "sparsity": 0.38154419263203937, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.v_proj": { | |
| "mse": 4.704533239419106e-06, | |
| "rmse": 0.0021689936005943188, | |
| "relative_error": 0.1397262341599038, | |
| "max_error": 0.0985720157623291, | |
| "sparsity": 0.3873201509316762, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.q_proj": { | |
| "mse": 5.126988526171772e-06, | |
| "rmse": 0.0022642854339000136, | |
| "relative_error": 0.1006711422581576, | |
| "max_error": 0.06201744079589844, | |
| "sparsity": 0.3899892270565033, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.encoder_attn.out_proj": { | |
| "mse": 4.576470018946566e-06, | |
| "rmse": 0.0021392685710182736, | |
| "relative_error": 0.14740875028747533, | |
| "max_error": 0.05412483215332031, | |
| "sparsity": 0.36713724335034686, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.fc1": { | |
| "mse": 4.29202964369324e-06, | |
| "rmse": 0.002071721420387703, | |
| "relative_error": 0.10361871269479109, | |
| "max_error": 0.18188095092773438, | |
| "sparsity": 0.39315704504648846, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.15.fc2": { | |
| "mse": 2.6174484446528368e-06, | |
| "rmse": 0.0016178530355544774, | |
| "relative_error": 0.11655210409825213, | |
| "max_error": 0.10709381103515625, | |
| "sparsity": 0.39326457182566327, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.self_attn.k_proj": { | |
| "mse": 7.54423808757565e-06, | |
| "rmse": 0.002746677645370066, | |
| "relative_error": 0.09956229625992792, | |
| "max_error": 0.11260986328125, | |
| "sparsity": 0.39282022913297016, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.self_attn.v_proj": { | |
| "mse": 2.7785231395682786e-06, | |
| "rmse": 0.0016668902602055957, | |
| "relative_error": 0.10057676252620366, | |
| "max_error": 0.034053802490234375, | |
| "sparsity": 0.3960314989089966, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.self_attn.q_proj": { | |
| "mse": 7.401999937428627e-06, | |
| "rmse": 0.0027206616727238663, | |
| "relative_error": 0.10152884904222975, | |
| "max_error": 0.09681415557861328, | |
| "sparsity": 0.39324483275413513, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.self_attn.out_proj": { | |
| "mse": 4.540847839962225e-06, | |
| "rmse": 0.002130926521483607, | |
| "relative_error": 0.13036370979166703, | |
| "max_error": 0.0471343994140625, | |
| "sparsity": 0.37679870923360187, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.k_proj": { | |
| "mse": 9.028391104948241e-06, | |
| "rmse": 0.0030047281249637615, | |
| "relative_error": 0.1303974522580755, | |
| "max_error": 0.10811614990234375, | |
| "sparsity": 0.3811313807964325, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.v_proj": { | |
| "mse": 6.043881967343623e-06, | |
| "rmse": 0.0024584307936860094, | |
| "relative_error": 0.1375850067512669, | |
| "max_error": 0.09338951110839844, | |
| "sparsity": 0.38720113039016724, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.q_proj": { | |
| "mse": 6.067926278774394e-06, | |
| "rmse": 0.002463316114260286, | |
| "relative_error": 0.10966318214631557, | |
| "max_error": 0.07547116279602051, | |
| "sparsity": 0.3893355329831441, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.encoder_attn.out_proj": { | |
| "mse": 4.7658741095801815e-06, | |
| "rmse": 0.0021830882047183028, | |
| "relative_error": 0.14874350560016716, | |
| "max_error": 0.15030765533447266, | |
| "sparsity": 0.36898477872212726, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.fc1": { | |
| "mse": 4.437824827618897e-06, | |
| "rmse": 0.002106614541775238, | |
| "relative_error": 0.1053240968561272, | |
| "max_error": 0.14221620559692383, | |
| "sparsity": 0.39214274287223816, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.16.fc2": { | |
| "mse": 2.5879278382490156e-06, | |
| "rmse": 0.0016087037757924905, | |
| "relative_error": 0.11553459435937694, | |
| "max_error": 0.09499549865722656, | |
| "sparsity": 0.3925200899442037, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.self_attn.k_proj": { | |
| "mse": 7.111822924343869e-06, | |
| "rmse": 0.002666800128308057, | |
| "relative_error": 0.10009080356704801, | |
| "max_error": 0.08137226104736328, | |
| "sparsity": 0.3919958472251892, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.self_attn.v_proj": { | |
| "mse": 2.804535597533686e-06, | |
| "rmse": 0.0016746747736601534, | |
| "relative_error": 0.10108355429673305, | |
| "max_error": 0.04348468780517578, | |
| "sparsity": 0.3957781990369161, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.self_attn.q_proj": { | |
| "mse": 6.8451031438598875e-06, | |
| "rmse": 0.002616314802132933, | |
| "relative_error": 0.0997202524197124, | |
| "max_error": 0.07852411270141602, | |
| "sparsity": 0.39373981952667236, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.self_attn.out_proj": { | |
| "mse": 5.031957243772922e-06, | |
| "rmse": 0.0022432024526941215, | |
| "relative_error": 0.13686692940105935, | |
| "max_error": 0.09112894535064697, | |
| "sparsity": 0.3731420934200287, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.k_proj": { | |
| "mse": 8.465343853458762e-06, | |
| "rmse": 0.0029095263967626695, | |
| "relative_error": 0.1292488277558289, | |
| "max_error": 0.09967803955078125, | |
| "sparsity": 0.3791249593098958, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.v_proj": { | |
| "mse": 5.18366277901805e-06, | |
| "rmse": 0.002276765859507308, | |
| "relative_error": 0.13927888064018468, | |
| "max_error": 0.12140274047851562, | |
| "sparsity": 0.3882832924524943, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.q_proj": { | |
| "mse": 5.289134605845902e-06, | |
| "rmse": 0.0022998118631413966, | |
| "relative_error": 0.10185402517236604, | |
| "max_error": 0.08112192153930664, | |
| "sparsity": 0.39048442244529724, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.encoder_attn.out_proj": { | |
| "mse": 5.164796675671823e-06, | |
| "rmse": 0.0022726189024277308, | |
| "relative_error": 0.15464162757039746, | |
| "max_error": 0.05605512857437134, | |
| "sparsity": 0.36702677607536316, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.fc1": { | |
| "mse": 4.299927240936086e-06, | |
| "rmse": 0.0020736265914903977, | |
| "relative_error": 0.10511625170131214, | |
| "max_error": 0.18689966201782227, | |
| "sparsity": 0.391550749540329, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.17.fc2": { | |
| "mse": 2.225296839242219e-06, | |
| "rmse": 0.0014917428864392882, | |
| "relative_error": 0.10765547812018275, | |
| "max_error": 0.07597094774246216, | |
| "sparsity": 0.3949810266494751, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.self_attn.k_proj": { | |
| "mse": 7.060702955641318e-06, | |
| "rmse": 0.0026571983282475016, | |
| "relative_error": 0.10042069897007623, | |
| "max_error": 0.1058502197265625, | |
| "sparsity": 0.3925777077674866, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.self_attn.v_proj": { | |
| "mse": 2.776995643216651e-06, | |
| "rmse": 0.001666432009779172, | |
| "relative_error": 0.1006445525288923, | |
| "max_error": 0.047902047634124756, | |
| "sparsity": 0.3957265118757884, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.self_attn.q_proj": { | |
| "mse": 7.198506409622496e-06, | |
| "rmse": 0.00268300324443011, | |
| "relative_error": 0.10175985238792434, | |
| "max_error": 0.09590697288513184, | |
| "sparsity": 0.39333292841911316, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.self_attn.out_proj": { | |
| "mse": 5.15109286425286e-06, | |
| "rmse": 0.0022696019175734014, | |
| "relative_error": 0.13814324086242794, | |
| "max_error": 0.05695158243179321, | |
| "sparsity": 0.37223102649052936, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.k_proj": { | |
| "mse": 9.266636880056467e-06, | |
| "rmse": 0.0030441151226680745, | |
| "relative_error": 0.1299368621722594, | |
| "max_error": 0.10103487968444824, | |
| "sparsity": 0.3819492657979329, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.v_proj": { | |
| "mse": 5.916096142755123e-06, | |
| "rmse": 0.0024323026420976323, | |
| "relative_error": 0.1364302537708723, | |
| "max_error": 0.13587188720703125, | |
| "sparsity": 0.3874334692955017, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.q_proj": { | |
| "mse": 5.99771146880812e-06, | |
| "rmse": 0.0024490225537565228, | |
| "relative_error": 0.11000250569288435, | |
| "max_error": 0.08011209964752197, | |
| "sparsity": 0.39144714673360187, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.encoder_attn.out_proj": { | |
| "mse": 5.144750048202695e-06, | |
| "rmse": 0.002268204146059762, | |
| "relative_error": 0.1501705195577681, | |
| "max_error": 0.07988643646240234, | |
| "sparsity": 0.3683264156182607, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.fc1": { | |
| "mse": 4.302585239202017e-06, | |
| "rmse": 0.0020742673981919537, | |
| "relative_error": 0.10558999669932816, | |
| "max_error": 0.09682464599609375, | |
| "sparsity": 0.39112422863642377, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.18.fc2": { | |
| "mse": 2.21805657929508e-06, | |
| "rmse": 0.001489314130496008, | |
| "relative_error": 0.10615648552835152, | |
| "max_error": 0.11030066013336182, | |
| "sparsity": 0.3956604599952698, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.self_attn.k_proj": { | |
| "mse": 7.654371074750088e-06, | |
| "rmse": 0.002766653407051575, | |
| "relative_error": 0.09998905693533489, | |
| "max_error": 0.10614013671875, | |
| "sparsity": 0.3909987409909566, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.self_attn.v_proj": { | |
| "mse": 2.9416580673569115e-06, | |
| "rmse": 0.0017151262540573832, | |
| "relative_error": 0.10296090184585628, | |
| "max_error": 0.07405853271484375, | |
| "sparsity": 0.3954276541868846, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.self_attn.q_proj": { | |
| "mse": 7.1339186433760915e-06, | |
| "rmse": 0.00267093965551004, | |
| "relative_error": 0.10066231041709936, | |
| "max_error": 0.08835959434509277, | |
| "sparsity": 0.392776886622111, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.self_attn.out_proj": { | |
| "mse": 5.691019850928569e-06, | |
| "rmse": 0.0023855858506724443, | |
| "relative_error": 0.14365555586736534, | |
| "max_error": 0.10810613632202148, | |
| "sparsity": 0.371090700229009, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.k_proj": { | |
| "mse": 9.113067790167406e-06, | |
| "rmse": 0.003018785813893958, | |
| "relative_error": 0.13066388863191528, | |
| "max_error": 0.10552120208740234, | |
| "sparsity": 0.3823689818382263, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.v_proj": { | |
| "mse": 6.058616691007046e-06, | |
| "rmse": 0.002461425743549264, | |
| "relative_error": 0.13716445381256673, | |
| "max_error": 0.11244583129882812, | |
| "sparsity": 0.38700806101163227, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.q_proj": { | |
| "mse": 5.2065956879232544e-06, | |
| "rmse": 0.002281796592144719, | |
| "relative_error": 0.1024064949886521, | |
| "max_error": 0.06893253326416016, | |
| "sparsity": 0.3919488489627838, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.encoder_attn.out_proj": { | |
| "mse": 5.72732960790745e-06, | |
| "rmse": 0.002393183989564415, | |
| "relative_error": 0.15476802542259985, | |
| "max_error": 0.143707275390625, | |
| "sparsity": 0.36928629875183105, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.fc1": { | |
| "mse": 4.3190607357246336e-06, | |
| "rmse": 0.002078235004931982, | |
| "relative_error": 0.1054971112165872, | |
| "max_error": 0.17518138885498047, | |
| "sparsity": 0.39132146040598553, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.19.fc2": { | |
| "mse": 2.1109028693899745e-06, | |
| "rmse": 0.001452894651855383, | |
| "relative_error": 0.10308541448904392, | |
| "max_error": 0.09531521797180176, | |
| "sparsity": 0.39622411131858826, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.self_attn.k_proj": { | |
| "mse": 6.925800335011445e-06, | |
| "rmse": 0.002631691534927953, | |
| "relative_error": 0.10031893636480439, | |
| "max_error": 0.09749603271484375, | |
| "sparsity": 0.3923425277074178, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.self_attn.v_proj": { | |
| "mse": 3.0535427413269645e-06, | |
| "rmse": 0.0017474389091830834, | |
| "relative_error": 0.1044799428844684, | |
| "max_error": 0.05923652648925781, | |
| "sparsity": 0.39478006958961487, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.self_attn.q_proj": { | |
| "mse": 7.033928795863176e-06, | |
| "rmse": 0.002652155499940223, | |
| "relative_error": 0.1007410236001971, | |
| "max_error": 0.097442626953125, | |
| "sparsity": 0.3926045795281728, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.self_attn.out_proj": { | |
| "mse": 6.458812094933819e-06, | |
| "rmse": 0.002541419307185223, | |
| "relative_error": 0.15222648411156842, | |
| "max_error": 0.11455154418945312, | |
| "sparsity": 0.36696451902389526, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.k_proj": { | |
| "mse": 9.325013706984464e-06, | |
| "rmse": 0.003053688541253751, | |
| "relative_error": 0.13199942726675934, | |
| "max_error": 0.09883594512939453, | |
| "sparsity": 0.3818304439385732, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.v_proj": { | |
| "mse": 6.201610176503891e-06, | |
| "rmse": 0.002490303229830434, | |
| "relative_error": 0.13482933944961417, | |
| "max_error": 0.12136650085449219, | |
| "sparsity": 0.38701191544532776, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.q_proj": { | |
| "mse": 6.560383098985767e-06, | |
| "rmse": 0.0025613244813935164, | |
| "relative_error": 0.11500719472949472, | |
| "max_error": 0.116424560546875, | |
| "sparsity": 0.39206013083457947, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.encoder_attn.out_proj": { | |
| "mse": 5.8767932387127075e-06, | |
| "rmse": 0.0024242098173864217, | |
| "relative_error": 0.15687287455120158, | |
| "max_error": 0.0757455825805664, | |
| "sparsity": 0.3680751621723175, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.fc1": { | |
| "mse": 4.304219146433752e-06, | |
| "rmse": 0.002074661212447409, | |
| "relative_error": 0.10639820632119394, | |
| "max_error": 0.1877884864807129, | |
| "sparsity": 0.39101404945055646, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.20.fc2": { | |
| "mse": 1.94695917343779e-06, | |
| "rmse": 0.0013953347890158082, | |
| "relative_error": 0.09960082322316022, | |
| "max_error": 0.0478672981262207, | |
| "sparsity": 0.3978532950083415, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.self_attn.k_proj": { | |
| "mse": 7.110678780009039e-06, | |
| "rmse": 0.0026665856033529166, | |
| "relative_error": 0.10068632857397676, | |
| "max_error": 0.08507537841796875, | |
| "sparsity": 0.39192261298497516, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.self_attn.v_proj": { | |
| "mse": 2.9248951705085346e-06, | |
| "rmse": 0.00171023249019206, | |
| "relative_error": 0.10292403303235359, | |
| "max_error": 0.06364059448242188, | |
| "sparsity": 0.39528748393058777, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.self_attn.q_proj": { | |
| "mse": 7.177129191404674e-06, | |
| "rmse": 0.0026790164597114132, | |
| "relative_error": 0.10230457563954339, | |
| "max_error": 0.09338188171386719, | |
| "sparsity": 0.39309224486351013, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.self_attn.out_proj": { | |
| "mse": 5.525982487597503e-06, | |
| "rmse": 0.0023507408380333003, | |
| "relative_error": 0.14147954622060568, | |
| "max_error": 0.06934088468551636, | |
| "sparsity": 0.3728704849878947, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.k_proj": { | |
| "mse": 9.634692105464637e-06, | |
| "rmse": 0.0031039800426975422, | |
| "relative_error": 0.12988385988091886, | |
| "max_error": 0.10436439514160156, | |
| "sparsity": 0.3817842702070872, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.v_proj": { | |
| "mse": 5.140742814546684e-06, | |
| "rmse": 0.0022673206245581333, | |
| "relative_error": 0.1396476253384051, | |
| "max_error": 0.11195755004882812, | |
| "sparsity": 0.3880261182785034, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.q_proj": { | |
| "mse": 5.8377454479341395e-06, | |
| "rmse": 0.002416142679548155, | |
| "relative_error": 0.10332553866401363, | |
| "max_error": 0.05857086181640625, | |
| "sparsity": 0.39286743601163227, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.encoder_attn.out_proj": { | |
| "mse": 4.931306648359168e-06, | |
| "rmse": 0.0022206545540356267, | |
| "relative_error": 0.14959711463987227, | |
| "max_error": 0.17577648162841797, | |
| "sparsity": 0.37272705634435016, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.fc1": { | |
| "mse": 4.3780514715763275e-06, | |
| "rmse": 0.0020923793804127223, | |
| "relative_error": 0.10695828960285728, | |
| "max_error": 0.1912097930908203, | |
| "sparsity": 0.3911544283231099, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.21.fc2": { | |
| "mse": 1.935687805598718e-06, | |
| "rmse": 0.001391289978975885, | |
| "relative_error": 0.09899131211097983, | |
| "max_error": 0.08098149299621582, | |
| "sparsity": 0.3980131546656291, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.self_attn.k_proj": { | |
| "mse": 7.28060240362538e-06, | |
| "rmse": 0.0026982591431560796, | |
| "relative_error": 0.10191833245112013, | |
| "max_error": 0.1451416015625, | |
| "sparsity": 0.39229288697242737, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.self_attn.v_proj": { | |
| "mse": 2.9060797714919318e-06, | |
| "rmse": 0.0017047227843529082, | |
| "relative_error": 0.10340695750489584, | |
| "max_error": 0.05385589599609375, | |
| "sparsity": 0.39453795552253723, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.self_attn.q_proj": { | |
| "mse": 7.23665971236187e-06, | |
| "rmse": 0.0026901040337432808, | |
| "relative_error": 0.1034717778998442, | |
| "max_error": 0.11018180847167969, | |
| "sparsity": 0.39337016145388287, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.self_attn.out_proj": { | |
| "mse": 5.837833214172861e-06, | |
| "rmse": 0.002416160841950068, | |
| "relative_error": 0.14739992509916447, | |
| "max_error": 0.13663291931152344, | |
| "sparsity": 0.36936360597610474, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.k_proj": { | |
| "mse": 8.946901289164089e-06, | |
| "rmse": 0.0029911371230961796, | |
| "relative_error": 0.1302047069315381, | |
| "max_error": 0.09290504455566406, | |
| "sparsity": 0.3810929258664449, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.v_proj": { | |
| "mse": 5.269354005577043e-06, | |
| "rmse": 0.0022955073525425798, | |
| "relative_error": 0.13953589220138896, | |
| "max_error": 0.08990097045898438, | |
| "sparsity": 0.3884460429350535, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.q_proj": { | |
| "mse": 5.811456958326744e-06, | |
| "rmse": 0.002410696363776812, | |
| "relative_error": 0.10386012010729784, | |
| "max_error": 0.06531906127929688, | |
| "sparsity": 0.39261190096537274, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.encoder_attn.out_proj": { | |
| "mse": 4.937660833093105e-06, | |
| "rmse": 0.0022220847943076128, | |
| "relative_error": 0.1467725544756901, | |
| "max_error": 0.09409737586975098, | |
| "sparsity": 0.3706894914309184, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.fc1": { | |
| "mse": 4.237566827214323e-06, | |
| "rmse": 0.002058535116828062, | |
| "relative_error": 0.10598931686848741, | |
| "max_error": 0.10584211349487305, | |
| "sparsity": 0.39103678862253827, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.22.fc2": { | |
| "mse": 1.9952328784711426e-06, | |
| "rmse": 0.001412527124862083, | |
| "relative_error": 0.0993138813309406, | |
| "max_error": 0.058658599853515625, | |
| "sparsity": 0.3975971539815267, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.self_attn.k_proj": { | |
| "mse": 7.5108710007043555e-06, | |
| "rmse": 0.0027405968329370077, | |
| "relative_error": 0.10329800215914882, | |
| "max_error": 0.11957883834838867, | |
| "sparsity": 0.392743319272995, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.self_attn.v_proj": { | |
| "mse": 2.932209554273868e-06, | |
| "rmse": 0.0017123695729234002, | |
| "relative_error": 0.10372649337928637, | |
| "max_error": 0.11333847045898438, | |
| "sparsity": 0.39437661568323773, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.self_attn.q_proj": { | |
| "mse": 7.364566954493057e-06, | |
| "rmse": 0.0027137735635997815, | |
| "relative_error": 0.10210883438700748, | |
| "max_error": 0.09737992286682129, | |
| "sparsity": 0.39333759744962055, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.self_attn.out_proj": { | |
| "mse": 5.632010470435489e-06, | |
| "rmse": 0.002373185721859014, | |
| "relative_error": 0.14341816756177525, | |
| "max_error": 0.1378253698348999, | |
| "sparsity": 0.3711596727371216, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.k_proj": { | |
| "mse": 9.51664969761623e-06, | |
| "rmse": 0.0030849067567134394, | |
| "relative_error": 0.13075529991499188, | |
| "max_error": 0.1080474853515625, | |
| "sparsity": 0.3818627993265788, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.v_proj": { | |
| "mse": 5.099964255350642e-06, | |
| "rmse": 0.0022583100441149885, | |
| "relative_error": 0.1362102196953864, | |
| "max_error": 0.08089375495910645, | |
| "sparsity": 0.3871195415655772, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.q_proj": { | |
| "mse": 5.843322924192762e-06, | |
| "rmse": 0.0024172966148556865, | |
| "relative_error": 0.1018798696262347, | |
| "max_error": 0.08249902725219727, | |
| "sparsity": 0.3913104236125946, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.encoder_attn.out_proj": { | |
| "mse": 5.30712941326783e-06, | |
| "rmse": 0.0023037207758901316, | |
| "relative_error": 0.1502554521872796, | |
| "max_error": 0.13018798828125, | |
| "sparsity": 0.3712678949038188, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.fc1": { | |
| "mse": 4.195026576780947e-06, | |
| "rmse": 0.00204817640274976, | |
| "relative_error": 0.1046047392061076, | |
| "max_error": 0.12958335876464844, | |
| "sparsity": 0.3912206490834554, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.23.fc2": { | |
| "mse": 2.077426870528143e-06, | |
| "rmse": 0.0014413281619839887, | |
| "relative_error": 0.10017117590499727, | |
| "max_error": 0.061614990234375, | |
| "sparsity": 0.39870961507161456, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.self_attn.k_proj": { | |
| "mse": 7.72945804783376e-06, | |
| "rmse": 0.0027801902898603468, | |
| "relative_error": 0.10109454684474335, | |
| "max_error": 0.13892364501953125, | |
| "sparsity": 0.393850713968277, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.self_attn.v_proj": { | |
| "mse": 2.904957455029944e-06, | |
| "rmse": 0.0017043935739816504, | |
| "relative_error": 0.10329595381321555, | |
| "max_error": 0.04791975021362305, | |
| "sparsity": 0.3954577644666036, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.self_attn.q_proj": { | |
| "mse": 7.429816832882352e-06, | |
| "rmse": 0.002725769035131618, | |
| "relative_error": 0.09977390169187604, | |
| "max_error": 0.06260108947753906, | |
| "sparsity": 0.39338521162668866, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.self_attn.out_proj": { | |
| "mse": 5.125093139213277e-06, | |
| "rmse": 0.0022638668554518125, | |
| "relative_error": 0.13679754752296824, | |
| "max_error": 0.074981689453125, | |
| "sparsity": 0.3751322329044342, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.k_proj": { | |
| "mse": 1.0251448657072615e-05, | |
| "rmse": 0.003201788352947867, | |
| "relative_error": 0.13121741321883937, | |
| "max_error": 0.10564041137695312, | |
| "sparsity": 0.38090963164965314, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.v_proj": { | |
| "mse": 4.671982424042653e-06, | |
| "rmse": 0.0021614769080521433, | |
| "relative_error": 0.13911473255425738, | |
| "max_error": 0.12041473388671875, | |
| "sparsity": 0.3883776863416036, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.q_proj": { | |
| "mse": 5.997791959089227e-06, | |
| "rmse": 0.0024490389868454987, | |
| "relative_error": 0.10207252376560083, | |
| "max_error": 0.06053638458251953, | |
| "sparsity": 0.39128051201502484, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.encoder_attn.out_proj": { | |
| "mse": 4.884094323642785e-06, | |
| "rmse": 0.0022099987157559128, | |
| "relative_error": 0.15097575816756378, | |
| "max_error": 0.05415630340576172, | |
| "sparsity": 0.3685123721758525, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.fc1": { | |
| "mse": 4.1780058381846175e-06, | |
| "rmse": 0.00204401708363326, | |
| "relative_error": 0.10385832939196378, | |
| "max_error": 0.24924850463867188, | |
| "sparsity": 0.39151788751284283, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.24.fc2": { | |
| "mse": 2.1364394342526793e-06, | |
| "rmse": 0.0014616564008865694, | |
| "relative_error": 0.09912129202488006, | |
| "max_error": 0.07556533813476562, | |
| "sparsity": 0.3984825114409129, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.self_attn.k_proj": { | |
| "mse": 7.1837484938441776e-06, | |
| "rmse": 0.0026802515728647893, | |
| "relative_error": 0.1008720765652819, | |
| "max_error": 0.08188819885253906, | |
| "sparsity": 0.39413594206174213, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.self_attn.v_proj": { | |
| "mse": 2.8171020858280826e-06, | |
| "rmse": 0.0016784224992021772, | |
| "relative_error": 0.10140862140541809, | |
| "max_error": 0.058258056640625, | |
| "sparsity": 0.3959297736485799, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.self_attn.q_proj": { | |
| "mse": 7.291312613233458e-06, | |
| "rmse": 0.002700243065583811, | |
| "relative_error": 0.10086810552825662, | |
| "max_error": 0.07587432861328125, | |
| "sparsity": 0.39403606454531354, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.self_attn.out_proj": { | |
| "mse": 4.715477643912891e-06, | |
| "rmse": 0.0021715150572613794, | |
| "relative_error": 0.13077725991345815, | |
| "max_error": 0.15969955921173096, | |
| "sparsity": 0.3786965012550354, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.k_proj": { | |
| "mse": 1.0018255125032738e-05, | |
| "rmse": 0.0031651627327884324, | |
| "relative_error": 0.12939450811589961, | |
| "max_error": 0.12250888347625732, | |
| "sparsity": 0.3817686041196187, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.v_proj": { | |
| "mse": 5.109501671540784e-06, | |
| "rmse": 0.0022604206846383226, | |
| "relative_error": 0.13932281989773324, | |
| "max_error": 0.0858154296875, | |
| "sparsity": 0.3888474504152934, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.q_proj": { | |
| "mse": 5.8599548538040835e-06, | |
| "rmse": 0.002420734362503264, | |
| "relative_error": 0.09970289467145303, | |
| "max_error": 0.05779671669006348, | |
| "sparsity": 0.39336730043093365, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.encoder_attn.out_proj": { | |
| "mse": 4.621794232662069e-06, | |
| "rmse": 0.002149835861795516, | |
| "relative_error": 0.13997831581953607, | |
| "max_error": 0.07830524444580078, | |
| "sparsity": 0.3724184234937032, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.fc1": { | |
| "mse": 4.0411237023363356e-06, | |
| "rmse": 0.00201025463619322, | |
| "relative_error": 0.10222836554627669, | |
| "max_error": 0.13097572326660156, | |
| "sparsity": 0.39175139864285785, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.25.fc2": { | |
| "mse": 2.28481712838402e-06, | |
| "rmse": 0.001511561156018512, | |
| "relative_error": 0.09843970686863819, | |
| "max_error": 0.10336685180664062, | |
| "sparsity": 0.3982831835746765, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.self_attn.k_proj": { | |
| "mse": 7.559828645753441e-06, | |
| "rmse": 0.0027495142563284595, | |
| "relative_error": 0.09878593287227969, | |
| "max_error": 0.08675050735473633, | |
| "sparsity": 0.3938686152299245, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.self_attn.v_proj": { | |
| "mse": 2.7506955575518077e-06, | |
| "rmse": 0.001658522100411028, | |
| "relative_error": 0.10109371994159619, | |
| "max_error": 0.047027587890625, | |
| "sparsity": 0.39503682653109234, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.self_attn.q_proj": { | |
| "mse": 7.100480161170708e-06, | |
| "rmse": 0.0026646726180097075, | |
| "relative_error": 0.09871422338029574, | |
| "max_error": 0.07471036911010742, | |
| "sparsity": 0.3939921061197917, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.self_attn.out_proj": { | |
| "mse": 4.8927827265288215e-06, | |
| "rmse": 0.002211963545479179, | |
| "relative_error": 0.13409655432579504, | |
| "max_error": 0.07868766784667969, | |
| "sparsity": 0.37614888946215314, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.k_proj": { | |
| "mse": 1.0031794772658031e-05, | |
| "rmse": 0.0031673008655096267, | |
| "relative_error": 0.1288299553635519, | |
| "max_error": 0.11860847473144531, | |
| "sparsity": 0.3809572358926137, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.v_proj": { | |
| "mse": 4.6288523662951775e-06, | |
| "rmse": 0.0021514767873010337, | |
| "relative_error": 0.13865595516749002, | |
| "max_error": 0.11671996116638184, | |
| "sparsity": 0.387549231449763, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.q_proj": { | |
| "mse": 5.812537892779801e-06, | |
| "rmse": 0.0024109205488318775, | |
| "relative_error": 0.10149536194614978, | |
| "max_error": 0.09484195709228516, | |
| "sparsity": 0.3931569556395213, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.encoder_attn.out_proj": { | |
| "mse": 4.706681465904694e-06, | |
| "rmse": 0.0021694887568053205, | |
| "relative_error": 0.15107708606145456, | |
| "max_error": 0.07437705993652344, | |
| "sparsity": 0.36915241678555805, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.fc1": { | |
| "mse": 4.146654191572452e-06, | |
| "rmse": 0.002036333516782664, | |
| "relative_error": 0.10329280666537241, | |
| "max_error": 0.25098633766174316, | |
| "sparsity": 0.39259542028109234, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.26.fc2": { | |
| "mse": 2.3110003439796856e-06, | |
| "rmse": 0.0015201974687453223, | |
| "relative_error": 0.09846057930692553, | |
| "max_error": 0.07316017150878906, | |
| "sparsity": 0.39834823211034137, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.self_attn.k_proj": { | |
| "mse": 7.749058568151668e-06, | |
| "rmse": 0.0027837130901283034, | |
| "relative_error": 0.10001740881903856, | |
| "max_error": 0.09989166259765625, | |
| "sparsity": 0.39306701223055523, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.self_attn.v_proj": { | |
| "mse": 2.9391637781372992e-06, | |
| "rmse": 0.00171439895535937, | |
| "relative_error": 0.10169791445046879, | |
| "max_error": 0.0809783935546875, | |
| "sparsity": 0.39528605341911316, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.self_attn.q_proj": { | |
| "mse": 7.271161393873626e-06, | |
| "rmse": 0.0026965091125144796, | |
| "relative_error": 0.10089499659668688, | |
| "max_error": 0.09291458129882812, | |
| "sparsity": 0.3933923343817393, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.self_attn.out_proj": { | |
| "mse": 4.367072051536525e-06, | |
| "rmse": 0.0020897540648450777, | |
| "relative_error": 0.1254672668354207, | |
| "max_error": 0.10006523132324219, | |
| "sparsity": 0.37970418731371564, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.k_proj": { | |
| "mse": 9.897006748360582e-06, | |
| "rmse": 0.0031459508496415804, | |
| "relative_error": 0.1273730833824496, | |
| "max_error": 0.10152530670166016, | |
| "sparsity": 0.38151692350705463, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.v_proj": { | |
| "mse": 6.550741545652272e-06, | |
| "rmse": 0.0025594416472450145, | |
| "relative_error": 0.13784227925920206, | |
| "max_error": 0.18347549438476562, | |
| "sparsity": 0.3888313869635264, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.q_proj": { | |
| "mse": 6.3485595092060976e-06, | |
| "rmse": 0.0025196347967922053, | |
| "relative_error": 0.10380759943647969, | |
| "max_error": 0.09928417205810547, | |
| "sparsity": 0.39255839586257935, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.encoder_attn.out_proj": { | |
| "mse": 8.796329893812072e-06, | |
| "rmse": 0.0029658607340554734, | |
| "relative_error": 0.17384918392027254, | |
| "max_error": 0.16146254539489746, | |
| "sparsity": 0.37418050567309064, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.fc1": { | |
| "mse": 4.111732323508477e-06, | |
| "rmse": 0.002027740694346414, | |
| "relative_error": 0.10297469235624884, | |
| "max_error": 0.18045806884765625, | |
| "sparsity": 0.392143855492274, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.27.fc2": { | |
| "mse": 2.3316322312894044e-06, | |
| "rmse": 0.0015269683137804151, | |
| "relative_error": 0.09770264337979621, | |
| "max_error": 0.09347295761108398, | |
| "sparsity": 0.3979528347651164, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.self_attn.k_proj": { | |
| "mse": 8.503747267241124e-06, | |
| "rmse": 0.002916118527639287, | |
| "relative_error": 0.09985664465805358, | |
| "max_error": 0.09625625610351562, | |
| "sparsity": 0.39266947905222577, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.self_attn.v_proj": { | |
| "mse": 3.070137836402864e-06, | |
| "rmse": 0.001752180880047167, | |
| "relative_error": 0.10268442457138177, | |
| "max_error": 0.04823708534240723, | |
| "sparsity": 0.3949188192685445, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.self_attn.q_proj": { | |
| "mse": 7.081852345436346e-06, | |
| "rmse": 0.0026611749933885117, | |
| "relative_error": 0.10118874653700154, | |
| "max_error": 0.07839775085449219, | |
| "sparsity": 0.3924410045146942, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.self_attn.out_proj": { | |
| "mse": 5.0970502343261614e-06, | |
| "rmse": 0.0022576647745682177, | |
| "relative_error": 0.13451187684815757, | |
| "max_error": 0.061634063720703125, | |
| "sparsity": 0.3754400610923767, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.k_proj": { | |
| "mse": 1.0878928151214495e-05, | |
| "rmse": 0.0032983220205453703, | |
| "relative_error": 0.12808165015409617, | |
| "max_error": 0.09624671936035156, | |
| "sparsity": 0.3812200923760732, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.v_proj": { | |
| "mse": 4.56340103482944e-06, | |
| "rmse": 0.0021362118422173022, | |
| "relative_error": 0.1371713520656024, | |
| "max_error": 0.08240413665771484, | |
| "sparsity": 0.389852911233902, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.q_proj": { | |
| "mse": 5.70409110878245e-06, | |
| "rmse": 0.0023883239120317096, | |
| "relative_error": 0.09743786509632198, | |
| "max_error": 0.054846763610839844, | |
| "sparsity": 0.39298970500628155, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.encoder_attn.out_proj": { | |
| "mse": 4.033752702525817e-06, | |
| "rmse": 0.002008420449638426, | |
| "relative_error": 0.13633681301337572, | |
| "max_error": 0.07563626766204834, | |
| "sparsity": 0.3738970955212911, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.fc1": { | |
| "mse": 4.19958359998418e-06, | |
| "rmse": 0.0020492885594723307, | |
| "relative_error": 0.10338519796475662, | |
| "max_error": 0.26813602447509766, | |
| "sparsity": 0.39146946867307025, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.28.fc2": { | |
| "mse": 2.226875722044497e-06, | |
| "rmse": 0.00149227200002027, | |
| "relative_error": 0.09459599267190874, | |
| "max_error": 0.0768423080444336, | |
| "sparsity": 0.3965551257133484, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.self_attn.k_proj": { | |
| "mse": 8.869013072398957e-06, | |
| "rmse": 0.0029780888288294823, | |
| "relative_error": 0.09961169883960393, | |
| "max_error": 0.1217193603515625, | |
| "sparsity": 0.3917592465877533, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.self_attn.v_proj": { | |
| "mse": 2.9806146812916268e-06, | |
| "rmse": 0.0017264456786391011, | |
| "relative_error": 0.10224876266734143, | |
| "max_error": 0.059680938720703125, | |
| "sparsity": 0.39422382911046344, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.self_attn.q_proj": { | |
| "mse": 6.473637313320069e-06, | |
| "rmse": 0.002544334355645906, | |
| "relative_error": 0.10289062364859558, | |
| "max_error": 0.10062217712402344, | |
| "sparsity": 0.39235392212867737, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.self_attn.out_proj": { | |
| "mse": 5.372486612031935e-06, | |
| "rmse": 0.0023178625093028997, | |
| "relative_error": 0.1405932498882154, | |
| "max_error": 0.06487178802490234, | |
| "sparsity": 0.37305683890978497, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.k_proj": { | |
| "mse": 1.1259038728894666e-05, | |
| "rmse": 0.003355449109865126, | |
| "relative_error": 0.12733843258541144, | |
| "max_error": 0.0995025634765625, | |
| "sparsity": 0.38166890541712445, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.v_proj": { | |
| "mse": 5.454831807583105e-06, | |
| "rmse": 0.002335558136202802, | |
| "relative_error": 0.1387764948109504, | |
| "max_error": 0.0793447494506836, | |
| "sparsity": 0.38814249634742737, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.q_proj": { | |
| "mse": 6.018663953000214e-06, | |
| "rmse": 0.0024532965481164756, | |
| "relative_error": 0.10165474258764118, | |
| "max_error": 0.06073951721191406, | |
| "sparsity": 0.3900683621565501, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.encoder_attn.out_proj": { | |
| "mse": 4.977966909791576e-06, | |
| "rmse": 0.002231135789187107, | |
| "relative_error": 0.1434446946606575, | |
| "max_error": 0.06038784980773926, | |
| "sparsity": 0.37243449687957764, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.fc1": { | |
| "mse": 4.312468718126183e-06, | |
| "rmse": 0.002076648433925729, | |
| "relative_error": 0.10318226776341134, | |
| "max_error": 0.15053653717041016, | |
| "sparsity": 0.3915984034538269, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.29.fc2": { | |
| "mse": 2.3233008050738135e-06, | |
| "rmse": 0.0015242377783908302, | |
| "relative_error": 0.09629652680392264, | |
| "max_error": 0.08018112182617188, | |
| "sparsity": 0.39694616198539734, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.self_attn.k_proj": { | |
| "mse": 8.88339400262339e-06, | |
| "rmse": 0.002980502307099156, | |
| "relative_error": 0.09883244829625779, | |
| "max_error": 0.09085321426391602, | |
| "sparsity": 0.3919799824555715, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.self_attn.v_proj": { | |
| "mse": 2.9065095077385195e-06, | |
| "rmse": 0.0017048488225465974, | |
| "relative_error": 0.10188573840496354, | |
| "max_error": 0.07731056213378906, | |
| "sparsity": 0.3933894832928975, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.self_attn.q_proj": { | |
| "mse": 6.8295021264930256e-06, | |
| "rmse": 0.0026133316143369606, | |
| "relative_error": 0.1014569241992705, | |
| "max_error": 0.10349559783935547, | |
| "sparsity": 0.39104756712913513, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.self_attn.out_proj": { | |
| "mse": 5.044690169597743e-06, | |
| "rmse": 0.002246038772950668, | |
| "relative_error": 0.1365998044489397, | |
| "max_error": 0.12729167938232422, | |
| "sparsity": 0.3748050928115845, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.k_proj": { | |
| "mse": 1.2722019164357334e-05, | |
| "rmse": 0.0035667939615791288, | |
| "relative_error": 0.13118177825127014, | |
| "max_error": 0.12677371501922607, | |
| "sparsity": 0.38109782338142395, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.v_proj": { | |
| "mse": 6.368978574755602e-06, | |
| "rmse": 0.00252368353300401, | |
| "relative_error": 0.13948196380849742, | |
| "max_error": 0.12861919403076172, | |
| "sparsity": 0.3880590796470642, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.q_proj": { | |
| "mse": 6.026943992765155e-06, | |
| "rmse": 0.0024549835015260603, | |
| "relative_error": 0.1013362306717025, | |
| "max_error": 0.06700611114501953, | |
| "sparsity": 0.3905596931775411, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.encoder_attn.out_proj": { | |
| "mse": 7.038618150545517e-06, | |
| "rmse": 0.0026530394174503922, | |
| "relative_error": 0.16141626771219036, | |
| "max_error": 0.12373542785644531, | |
| "sparsity": 0.3692755202452342, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.fc1": { | |
| "mse": 4.4825660552305635e-06, | |
| "rmse": 0.002117207135646053, | |
| "relative_error": 0.1025177058122054, | |
| "max_error": 0.10970878601074219, | |
| "sparsity": 0.39209650953610736, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.30.fc2": { | |
| "mse": 2.866476506824256e-06, | |
| "rmse": 0.0016930671891051035, | |
| "relative_error": 0.10729600679037879, | |
| "max_error": 0.12681865692138672, | |
| "sparsity": 0.39671417077382404, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.self_attn.k_proj": { | |
| "mse": 9.31453632802004e-06, | |
| "rmse": 0.003051972530679141, | |
| "relative_error": 0.09897321660785756, | |
| "max_error": 0.10159587860107422, | |
| "sparsity": 0.3901355067888896, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.self_attn.v_proj": { | |
| "mse": 2.9529519451898523e-06, | |
| "rmse": 0.001718415533330007, | |
| "relative_error": 0.10543492738240678, | |
| "max_error": 0.06787973642349243, | |
| "sparsity": 0.3939440945784251, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.self_attn.q_proj": { | |
| "mse": 6.700065114273457e-06, | |
| "rmse": 0.0025884483989976424, | |
| "relative_error": 0.10306689057486865, | |
| "max_error": 0.07091546058654785, | |
| "sparsity": 0.3918269872665405, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.self_attn.out_proj": { | |
| "mse": 5.729709300794639e-06, | |
| "rmse": 0.002393681119279391, | |
| "relative_error": 0.14882814341736078, | |
| "max_error": 0.141998291015625, | |
| "sparsity": 0.37238483627637226, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.k_proj": { | |
| "mse": 1.6164902262971736e-05, | |
| "rmse": 0.004020559943959515, | |
| "relative_error": 0.13258171942511154, | |
| "max_error": 0.13735008239746094, | |
| "sparsity": 0.3815281093120575, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.v_proj": { | |
| "mse": 8.979875929071568e-06, | |
| "rmse": 0.0029966441111802995, | |
| "relative_error": 0.14800653019494162, | |
| "max_error": 0.15266036987304688, | |
| "sparsity": 0.3868330915768941, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.q_proj": { | |
| "mse": 7.939345778140705e-06, | |
| "rmse": 0.002817684471004641, | |
| "relative_error": 0.10720343656915154, | |
| "max_error": 0.097991943359375, | |
| "sparsity": 0.38758830229441327, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.encoder_attn.out_proj": { | |
| "mse": 8.937723578128498e-06, | |
| "rmse": 0.0029896025786262125, | |
| "relative_error": 0.1689557080065116, | |
| "max_error": 0.16315460205078125, | |
| "sparsity": 0.36957783500353497, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.fc1": { | |
| "mse": 5.8857654039456975e-06, | |
| "rmse": 0.0024260596455869953, | |
| "relative_error": 0.10615411099343097, | |
| "max_error": 0.13378000259399414, | |
| "sparsity": 0.3928763270378113, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.decoder.layers.31.fc2": { | |
| "mse": 3.943322553823236e-06, | |
| "rmse": 0.001985780086974194, | |
| "relative_error": 0.13137963639662306, | |
| "max_error": 0.224822998046875, | |
| "sparsity": 0.3983573913574219, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| } | |
| }, | |
| "summary": { | |
| "method_name": "Broad-tritplane3", | |
| "model_family": "seq2seq_lm", | |
| "selected_components": [ | |
| "decoder" | |
| ], | |
| "quantized_modules": 320, | |
| "quantized_params": 838860800, | |
| "quantized_fraction": 0.5434829481561585, | |
| "avg_relative_error": 0.12498197247780571, | |
| "avg_effective_bits": 9.0, | |
| "full_model_effective_bits": 12.195619362906891, | |
| "compression_ratio": 1.3119464886436334 | |
| }, | |
| "method_name": "Broad-tritplane3", | |
| "model_family": "seq2seq_lm", | |
| "format_family": "tritplane_small", | |
| "format_version": "1.0", | |
| "total_packed_bytes": 943718400, | |
| "total_fp16_bytes": 1677721600, | |
| "compression_ratio": 1.7777777777777777 | |
| } |