PolyPythias
- Preview • Updated • 85
EleutherAI/pile-preshuffled-seeds
Updated • 264 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 14.8k • 29 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 2.27k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 1.91k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 1.79k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 1.74k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 2.05k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 1.93k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 1.96k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 1.98k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 1.98k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 1.62k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 1.71k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 1.69k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 1.65k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 1.67k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 1.67k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 1.7k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 1.68k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 1.63k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 1.69k -
EleutherAI/pythia-70m
95.6M • Updated • 579k • 81 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 4.66k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 4.77k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 4.65k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 4.66k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 4.43k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 4.5k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 4.22k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 4.19k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 4.22k -
EleutherAI/pythia-160m
Text Generation • 0.2B • Updated • 3.02M • 42 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 6.84k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 6.08k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 6.38k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 3.93k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 3.54k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 3.47k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 3.54k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 3.53k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 3.49k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 131k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 3.7k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 3.85k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 3.41k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 3.13k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 2.92k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 2.89k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 2.96k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 2.83k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 2.92k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 1.17kNote 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 • 1.18kNote 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 • 1.17kNote 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 • 1.17kNote 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 • 1.16kNote 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 • 1.17kNote 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