{
  "best_metric": 0.8378005820186859,
  "best_model_checkpoint": "./hatebert_finetuned_v5/checkpoint-5000",
  "epoch": 0.7519739315703722,
  "eval_steps": 1000,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06266449429753101,
      "grad_norm": 2.7858245372772217,
      "learning_rate": 4.160401002506266e-06,
      "loss": 0.5658,
      "step": 500
    },
    {
      "epoch": 0.12532898859506203,
      "grad_norm": 0.6836428642272949,
      "learning_rate": 8.320802005012531e-06,
      "loss": 0.3394,
      "step": 1000
    },
    {
      "epoch": 0.12532898859506203,
      "eval_accuracy": 0.9509008303305656,
      "eval_f1": 0.7696604439217992,
      "eval_loss": 0.3321492373943329,
      "eval_optimal_threshold": 0.7856788039207458,
      "eval_runtime": 24.5664,
      "eval_samples_per_second": 1299.13,
      "eval_steps_per_second": 40.625,
      "step": 1000
    },
    {
      "epoch": 0.18799348289259304,
      "grad_norm": 0.07630700618028641,
      "learning_rate": 1.2497911445279867e-05,
      "loss": 0.3098,
      "step": 1500
    },
    {
      "epoch": 0.25065797719012406,
      "grad_norm": 0.6884164214134216,
      "learning_rate": 1.66750208855472e-05,
      "loss": 0.3245,
      "step": 2000
    },
    {
      "epoch": 0.25065797719012406,
      "eval_accuracy": 0.9620241265862447,
      "eval_f1": 0.8181272509003601,
      "eval_loss": 0.24252600967884064,
      "eval_optimal_threshold": 0.6846858263015747,
      "eval_runtime": 24.5401,
      "eval_samples_per_second": 1300.526,
      "eval_steps_per_second": 40.668,
      "step": 2000
    },
    {
      "epoch": 0.3133224714876551,
      "grad_norm": 1.6269655227661133,
      "learning_rate": 1.9906234043540826e-05,
      "loss": 0.2939,
      "step": 2500
    },
    {
      "epoch": 0.3759869657851861,
      "grad_norm": 53.418663024902344,
      "learning_rate": 1.9442046140277588e-05,
      "loss": 0.2911,
      "step": 3000
    },
    {
      "epoch": 0.3759869657851861,
      "eval_accuracy": 0.9669434435218549,
      "eval_f1": 0.8339889850511408,
      "eval_loss": 0.32369741797447205,
      "eval_optimal_threshold": 0.639145016670227,
      "eval_runtime": 24.5157,
      "eval_samples_per_second": 1301.819,
      "eval_steps_per_second": 40.709,
      "step": 3000
    },
    {
      "epoch": 0.43865146008271716,
      "grad_norm": 2.552913188934326,
      "learning_rate": 1.8977858237014345e-05,
      "loss": 0.2717,
      "step": 3500
    },
    {
      "epoch": 0.5013159543802481,
      "grad_norm": 74.53817749023438,
      "learning_rate": 1.8513670333751103e-05,
      "loss": 0.2574,
      "step": 4000
    },
    {
      "epoch": 0.5013159543802481,
      "eval_accuracy": 0.9656587811373962,
      "eval_f1": 0.8311768330252619,
      "eval_loss": 0.344524085521698,
      "eval_optimal_threshold": 0.903037965297699,
      "eval_runtime": 24.5127,
      "eval_samples_per_second": 1301.981,
      "eval_steps_per_second": 40.714,
      "step": 4000
    },
    {
      "epoch": 0.5639804486777792,
      "grad_norm": 4.5688018798828125,
      "learning_rate": 1.8049482430487864e-05,
      "loss": 0.2455,
      "step": 4500
    },
    {
      "epoch": 0.6266449429753101,
      "grad_norm": 0.18218272924423218,
      "learning_rate": 1.7585294527224622e-05,
      "loss": 0.278,
      "step": 5000
    },
    {
      "epoch": 0.6266449429753101,
      "eval_accuracy": 0.966818110606298,
      "eval_f1": 0.8378005820186859,
      "eval_loss": 0.22682912647724152,
      "eval_optimal_threshold": 0.8996956944465637,
      "eval_runtime": 24.5217,
      "eval_samples_per_second": 1301.5,
      "eval_steps_per_second": 40.699,
      "step": 5000
    },
    {
      "epoch": 0.6893094372728412,
      "grad_norm": 1.167153239250183,
      "learning_rate": 1.712110662396138e-05,
      "loss": 0.2859,
      "step": 5500
    },
    {
      "epoch": 0.7519739315703722,
      "grad_norm": 8.231012344360352,
      "learning_rate": 1.665691872069814e-05,
      "loss": 0.2335,
      "step": 6000
    },
    {
      "epoch": 0.7519739315703722,
      "eval_accuracy": 0.9660661131129563,
      "eval_f1": 0.8360333080999243,
      "eval_loss": 0.2758038341999054,
      "eval_optimal_threshold": 0.3707093894481659,
      "eval_runtime": 24.5444,
      "eval_samples_per_second": 1300.296,
      "eval_steps_per_second": 40.661,
      "step": 6000
    }
  ],
  "logging_steps": 500,
  "max_steps": 23937,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}