PolyPythias - a EleutherAI Collection

EleutherAI 's Collections

Pythia Scaling Suite

Quirky Models and Datasets

Quirky Math (archive)

Pile-T5 T5x Checkpoints

Sparse Autoencoders

PolyPythias

updated Apr 9, 2025

EleutherAI/polypythias-evals

Preview • Updated Sep 11, 2024 • 52
EleutherAI/pile-preshuffled-seeds

Updated Feb 12 • 169 • 1

Note Training data information for each seed.
EleutherAI/pythia-14m-deduped

Text Generation • 39.2M • Updated Feb 12 • 49k • 28
EleutherAI/pythia-14m-seed1

Text Generation • Updated Jan 11 • 3.23k
EleutherAI/pythia-14m-seed2

Text Generation • Updated Jan 11 • 3.1k
EleutherAI/pythia-14m-seed3

Text Generation • Updated Jan 11 • 3.08k
EleutherAI/pythia-14m-seed4

Text Generation • Updated Jan 11 • 2.69k
EleutherAI/pythia-14m-seed5

Text Generation • Updated Jan 11 • 2.72k
EleutherAI/pythia-14m-seed6

Text Generation • Updated Jan 11 • 2.93k
EleutherAI/pythia-14m-seed7

Text Generation • Updated Jan 11 • 3.02k
EleutherAI/pythia-14m-seed8

Text Generation • Updated Jan 11 • 2.72k
EleutherAI/pythia-14m-seed9

Text Generation • Updated Jan 11 • 2.95k
EleutherAI/pythia-31m-deduped

Text Generation • 55.7M • Updated Feb 12 • 1.49k • 5
EleutherAI/pythia-31m-seed1

Text Generation • Updated Jan 11 • 1.68k
EleutherAI/pythia-31m-seed2

Text Generation • Updated Jan 11 • 1.71k
EleutherAI/pythia-31m-seed3

Text Generation • Updated Jan 11 • 1.95k
EleutherAI/pythia-31m-seed4

Text Generation • Updated Jan 11 • 1.34k
EleutherAI/pythia-31m-seed5

Text Generation • Updated Jan 11 • 1.6k
EleutherAI/pythia-31m-seed6

Text Generation • Updated Jan 11 • 1.53k
EleutherAI/pythia-31m-seed7

Text Generation • Updated Jan 11 • 1.57k
EleutherAI/pythia-31m-seed8

Text Generation • Updated Jan 11 • 1.31k
EleutherAI/pythia-31m-seed9

Text Generation • Updated Jan 11 • 1.78k
EleutherAI/pythia-70m

95.6M • Updated Nov 21, 2023 • 273k • 79
EleutherAI/pythia-70m-seed1

Text Generation • Updated Jan 11 • 1.86k
EleutherAI/pythia-70m-seed2

Text Generation • Updated Jan 11 • 1.83k
EleutherAI/pythia-70m-seed3

Text Generation • Updated Jan 11 • 1.94k
EleutherAI/pythia-70m-seed4

Text Generation • Updated Jan 11 • 1.45k
EleutherAI/pythia-70m-seed5

Text Generation • Updated Jan 11 • 1.62k
EleutherAI/pythia-70m-seed6

Text Generation • Updated Jan 11 • 1.38k
EleutherAI/pythia-70m-seed7

Text Generation • Updated Jan 11 • 1.56k
EleutherAI/pythia-70m-seed8

Text Generation • Updated Jan 11 • 1.55k
EleutherAI/pythia-70m-seed9

Text Generation • Updated Jan 11 • 1.36k
EleutherAI/pythia-160m

Text Generation • Updated Jul 9, 2023 • 2.85M • 39
EleutherAI/pythia-160m-seed1

Text Generation • 0.2B • Updated Jan 11 • 2.86k
EleutherAI/pythia-160m-seed2

Text Generation • 0.2B • Updated Jan 11 • 2.76k
EleutherAI/pythia-160m-seed3

Text Generation • 0.2B • Updated Jan 11 • 2.76k
EleutherAI/pythia-160m-seed4

Text Generation • Updated Jan 11 • 1.79k • 1
EleutherAI/pythia-160m-seed5

Text Generation • Updated Jan 11 • 1.63k
EleutherAI/pythia-160m-seed6

Text Generation • Updated Jan 11 • 1.39k
EleutherAI/pythia-160m-seed7

Text Generation • Updated Jan 11 • 1.67k
EleutherAI/pythia-160m-seed8

Text Generation • Updated Jan 11 • 1.37k
EleutherAI/pythia-160m-seed9

Text Generation • Updated Jan 11 • 1.49k
EleutherAI/pythia-410m

Text Generation • 0.5B • Updated Jul 9, 2023 • 115k • 36
EleutherAI/pythia-410m-seed1

Text Generation • Updated Jan 11 • 1.82k
EleutherAI/pythia-410m-seed2

Text Generation • Updated Jan 11 • 2.03k
EleutherAI/pythia-410m-seed3

Text Generation • Updated Jan 11 • 1.84k
EleutherAI/pythia-410m-seed4

Text Generation • Updated Jan 11 • 1.39k
EleutherAI/pythia-410m-seed5

Text Generation • Updated Jan 11 • 1.37k
EleutherAI/pythia-410m-seed6

Text Generation • Updated Jan 11 • 1.39k • 1
EleutherAI/pythia-410m-seed7

Text Generation • Updated Jan 11 • 1.41k
EleutherAI/pythia-410m-seed8

Text Generation • Updated Jan 11 • 1.35k
EleutherAI/pythia-410m-seed9

Text Generation • Updated Jan 11 • 1.4k
EleutherAI/pythia-160m-data-seed1

Text Generation • Updated Jan 11 • 542

Note 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 Jan 11 • 909

Note 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 Jan 11 • 723

Note 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 Jan 11 • 312

Note 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 Jan 11 • 268

Note 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 Jan 11 • 249

Note 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 Mar 12, 2025