PolyPythias
- Preview • Updated • 62
EleutherAI/pile-preshuffled-seeds
Updated • 230 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 7.52k • 29 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 2.12k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 2.24k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 2.14k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 2.3k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 2.57k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 2.27k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 2.13k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 2.3k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 2.42k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 3.99k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 3.23k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 2.99k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 3.13k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 2.3k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 2.79k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 2.7k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 3.06k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 2.29k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 3.13k -
EleutherAI/pythia-70m
95.6M • Updated • 445k • 79 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 3.61k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 2.99k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 3.04k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 2.82k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 2.85k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 2.81k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 3.56k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 3.51k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 2.89k -
EleutherAI/pythia-160m
Text Generation • Updated • 2.67M • 39 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 5.5k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 5.36k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 4.99k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 3.76k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 4.2k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 3.45k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 3.93k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 3.46k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 3.85k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 172k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 6.6k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 3.05k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 3.34k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 3.1k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 3.01k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 2.87k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 2.85k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 2.79k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 2.85k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 3.47kNote 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 • 3.03kNote 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 • 3.37kNote 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 • 2.63kNote 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 • 2.65kNote 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 • 2.65kNote 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