PolyPythias
- Preview • Updated • 50
EleutherAI/pile-preshuffled-seeds
Updated • 232 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 20.7k • 29 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 5.21k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 5.17k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 5.05k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 4.63k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 5k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 5.12k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 5.23k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 4.66k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 5.41k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 3.71k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 4.67k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 4.47k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 4.93k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 3.31k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 4.32k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 4.12k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 4.52k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 3.28k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 4.83k -
EleutherAI/pythia-70m
95.6M • Updated • 344k • 79 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 4.8k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 4.01k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 4.06k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 3.64k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 3.93k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 3.63k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 4.81k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 4.74k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 3.68k -
EleutherAI/pythia-160m
Text Generation • Updated • 2.55M • 39 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 6.22k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 6.21k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 6.05k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 5.13k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 5.53k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 4.18k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 5.34k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 4.16k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 4.34k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 194k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 5.84k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 3.92k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 4.04k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 3.73k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 3.72k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 3.58k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 3.63k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 3.57k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 3.61k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 3.76kNote 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 • 3.69kNote 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-data-seed3
Text Generation • Updated • 3.85kNote 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 • 2.42kNote 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 • 2.4kNote 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 • 2.42kNote 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