PolyPythias
- Preview • Updated • 106
EleutherAI/pile-preshuffled-seeds
Updated • 85 • 1Note Training data information for each seed.
EleutherAI/pythia-14m
Text Generation • Updated • 203k • 23EleutherAI/pythia-14m-seed1
Updated • 4.98kEleutherAI/pythia-14m-seed2
Updated • 833EleutherAI/pythia-14m-seed3
Updated • 811EleutherAI/pythia-14m-seed4
Updated • 814EleutherAI/pythia-14m-seed5
Updated • 820EleutherAI/pythia-14m-seed6
Updated • 41EleutherAI/pythia-14m-seed7
Updated • 41EleutherAI/pythia-14m-seed8
Updated • 41EleutherAI/pythia-14m-seed9
Updated • 41EleutherAI/pythia-31m
Text Generation • Updated • 49.3k • 5EleutherAI/pythia-31m-seed1
Updated • 2.48kEleutherAI/pythia-31m-seed2
Updated • 427EleutherAI/pythia-31m-seed3
Updated • 507EleutherAI/pythia-31m-seed4
Updated • 501EleutherAI/pythia-31m-seed5
Updated • 502EleutherAI/pythia-31m-seed6
Updated • 41EleutherAI/pythia-31m-seed7
Updated • 42EleutherAI/pythia-31m-seed8
Updated • 41EleutherAI/pythia-31m-seed9
Updated • 41EleutherAI/pythia-70m
Updated • 114k • 66EleutherAI/pythia-70m-seed1
Updated • 2.5kEleutherAI/pythia-70m-seed2
Updated • 507EleutherAI/pythia-70m-seed3
Updated • 507EleutherAI/pythia-70m-seed4
Updated • 504EleutherAI/pythia-70m-seed5
Updated • 508EleutherAI/pythia-70m-seed6
Updated • 41EleutherAI/pythia-70m-seed7
Updated • 41EleutherAI/pythia-70m-seed8
Updated • 41EleutherAI/pythia-70m-seed9
Updated • 41EleutherAI/pythia-160m
Text Generation • Updated • 161k • 31EleutherAI/pythia-160m-seed1
Text Generation • Updated • 2.59kEleutherAI/pythia-160m-seed2
Text Generation • Updated • 1.07kEleutherAI/pythia-160m-seed3
Text Generation • Updated • 1.06kEleutherAI/pythia-160m-seed4
Updated • 496 • 1EleutherAI/pythia-160m-seed5
Updated • 513EleutherAI/pythia-160m-seed6
Updated • 43EleutherAI/pythia-160m-seed7
Updated • 41EleutherAI/pythia-160m-seed8
Updated • 41EleutherAI/pythia-160m-seed9
Updated • 31EleutherAI/pythia-410m
Text Generation • Updated • 81.1k • 26EleutherAI/pythia-410m-seed1
Updated • 1.49kEleutherAI/pythia-410m-seed2
Updated • 193EleutherAI/pythia-410m-seed3
Updated • 198EleutherAI/pythia-410m-seed4
Updated • 193EleutherAI/pythia-410m-seed5
Updated • 196EleutherAI/pythia-410m-seed6
Updated • 168EleutherAI/pythia-410m-seed7
Updated • 197EleutherAI/pythia-410m-seed8
Updated • 169EleutherAI/pythia-410m-seed9
Updated • 197
EleutherAI/pythia-160m-data-seed1
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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