PolyPythias
- Preview • Updated • 57
EleutherAI/pile-preshuffled-seeds
Updated • 230 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 7.75k • 29 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 1.61k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 1.74k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 1.63k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 2.36k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 2.63k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 2.04k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 1.62k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 2.34k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 1.88k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 4k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 2.72k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 2.5k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 2.73k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 2.34k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 2.64k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 2.46k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 2.63k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 2.34k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 2.58k -
EleutherAI/pythia-70m
95.6M • Updated • 444k • 79 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 3.42k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 3.04k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 3.09k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 2.87k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 2.9k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 2.86k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 3.06k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 2.99k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 2.94k -
EleutherAI/pythia-160m
Text Generation • Updated • 2.76M • 39 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 5.65k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 5.5k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 5.09k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 3.54k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 3.67k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 3.48k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 3.4k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 3.49k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 3.89k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 165k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 6.62k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 3.1k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 3.45k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 3.12k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 3.17k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 3.03k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 3.03k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 2.99k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 3.02k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 2.9kNote 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 • 2.39kNote 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 • 2.74kNote 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 • 2.64kNote 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 • 2.67kNote 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