PolyPythias
- Preview • Updated • 797
EleutherAI/pile-preshuffled-seeds
Updated • 337 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 34k • 31 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 1.62k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 1.39k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 1.23k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 1.11k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 1.28k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 2.44k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 1.17k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 1.13k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 971 -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 2.46k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 921 -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 979 -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 805 -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 778 -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 795 -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 1.38k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 741 -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 723 -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 625 -
EleutherAI/pythia-70m
95.6M • Updated • 502k • 85 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 4.51k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 2.9k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 2.2k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 1.63k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 1.57k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 1.84k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 1.5k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 1.54k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 1.39k -
EleutherAI/pythia-160m
Text Generation • 0.2B • Updated • 3.42M • 45 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 11.5k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 8.08k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 21.5k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 2.43k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 2.27k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 2.62k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 1.96k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 1.89k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 1.73k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 217k • 40 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 17.9k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 5.72k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 4.03k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 2.6k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 2.22k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 1.72k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 1.86k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 1.66k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 1.61k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 9.33kNote 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 • 4.06kNote 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 • 4.02kNote 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 • 4.76kNote 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 • 4.1kNote 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 • 3.96kNote 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