PolyPythias
- Preview • Updated • 3.12k
EleutherAI/pile-preshuffled-seeds
Updated • 259 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 25.5k • 30 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 2.29k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 1.77k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 1.43k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 1.13k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 1.45k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 1.69k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 1.34k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 1.26k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 1.25k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 1.79k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 975 -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 973 -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 915 -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 827 -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 790 -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 839 -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 755 -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 716 -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 776 -
EleutherAI/pythia-70m
95.6M • Updated • 484k • 85 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 3.99k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 2.42k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 1.81k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 1.2k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 1.13k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 1.2k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 1.05k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 1.1k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 1.1k -
EleutherAI/pythia-160m
Text Generation • 0.2B • Updated • 3.49M • 45 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 39.4k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 36k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 30.7k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 15.1k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 15.2k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 15.1k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 3.94k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 3.74k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 4.21k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 208k • 40 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 19k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 6.78k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 4.15k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 2.89k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 2.93k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 3.5k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 4.77k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 5.46k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 6.13k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 51.6kNote 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 • 11.3kNote 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 • 7.69kNote 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 • 19.9kNote 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 • 12.7kNote 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 • 5.62kNote 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