PolyPythias
- Preview • Updated • 159
EleutherAI/pile-preshuffled-seeds
Updated • 556 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 11.4k • 29 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 1.56k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 1.5k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 1.43k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 1.39k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 1.82k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 2.39k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 2.01k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 2.06k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 1.92k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 2k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 1.94k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 2.09k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 1.9k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 2.07k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 1.85k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 1.93k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 1.83k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 1.88k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 1.97k -
EleutherAI/pythia-70m
95.6M • Updated • 548k • 82 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 2.54k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 2.4k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 2.1k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 1.95k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 1.7k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 1.6k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 1.52k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 1.75k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 1.5k -
EleutherAI/pythia-160m
Text Generation • 0.2B • Updated • 2.99M • 42 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 4.3k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 2.47k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 2.66k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 1.69k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 1.61k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 1.6k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 1.51k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 1.5k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 1.54k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 173k • 38 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 4.03k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 4.21k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 8.14k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 4.55k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 3.37k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 2.36k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 2.71k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 2.37k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 2.32k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 740Note Version where the data order and weight initialization seeds are decoupled. Here, only the data seed is different from pythia-160m ("seed 0").
EleutherAI/pythia-160m-data-seed2
Text Generation • Updated • 660Note Version where the data order and weight initialization seeds are decoupled. Here, only the data seed is different from pythia-160m ("seed 0").
EleutherAI/pythia-160m-data-seed3
Text Generation • Updated • 655Note Version where the data order and weight initialization seeds are decoupled. Here, only the data seed is different from pythia-160m ("seed 0").
EleutherAI/pythia-160m-weight-seed1
Text Generation • Updated • 734Note Version where the data order and weight initialization seeds are decoupled. Here, only the weight initialization seed is different from pythia-160m ("seed 0").
EleutherAI/pythia-160m-weight-seed2
Text Generation • Updated • 656Note Version where the data order and weight initialization seeds are decoupled. Here, only the weight initialization seed is different from pythia-160m ("seed 0").
EleutherAI/pythia-160m-weight-seed3
Text Generation • Updated • 663Note Version where the data order and weight initialization seeds are decoupled. Here, only the weight initialization seed is different from pythia-160m ("seed 0").
-
PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs
Paper • 2503.09543 • Published