PolyPythias
-
Preview • Updated • 600
EleutherAI/pile-preshuffled-seeds
Updated • 39 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m
Text Generation • 39.2M • Updated • 43.8k • 27 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 26 -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 160 -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 27 -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 30 -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 24 -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 24 -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 30 -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 31 -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 29 -
EleutherAI/pythia-31m
Text Generation • 55.7M • Updated • 3.81k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 29 -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 25 -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 29 -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 27 -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 26 -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 29 -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 25 -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 26 -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 28 -
EleutherAI/pythia-70m
95.6M • Updated • 48.9k • 77 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 1.31k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 2.47k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 2.76k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 40 -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 26 -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 27 -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 27 -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 26 -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 22 -
EleutherAI/pythia-160m
Text Generation • 0.2B • Updated • 74.4k • 36 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 259 -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 284 -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 232 -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 31 • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 28 -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 30 -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 29 -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 26 -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 26 -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 35k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 428 -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 333 -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 169 -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 252 -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 128 -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 125 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 130 -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 167 -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 161
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 34Note 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 • 29Note 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 • 23Note 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 • 28Note 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 • 28Note 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 • 27Note 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