PolyPythias
-
Preview • Updated • 94
EleutherAI/pile-preshuffled-seeds
Updated • 300 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 56.4k • 28 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 154 -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 225 -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 110 -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 103 -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 139 -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 127 -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 139 -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 134 -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 131 -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 3.06k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 190 -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 191 -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 99 -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 95 -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 102 -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 106 -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 104 -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 100 -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 103 -
EleutherAI/pythia-70m
95.6M • Updated • 83.9k • 78 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 947 -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 630 -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 565 -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 529 -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 552 -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 525 -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 554 -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 535 -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 521 -
EleutherAI/pythia-160m
Text Generation • Updated • 1.42M • 38 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 1.06k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 1.01k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 914 -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 846 • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 612 -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 596 -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 599 -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 599 -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 583 -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 72.4k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 660 -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 1.29k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 609 -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 568 -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 537 -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 846 • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 548 -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 571 -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 713
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 135Note 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 • 138Note 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 • 141Note 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 • 248Note 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 • 216Note 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 • 208Note 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