PolyPythias
- Preview • Updated • 53
EleutherAI/pile-preshuffled-seeds
Updated • 231 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 22.1k • 29 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 5.07k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 5.2k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 5.15k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 4.5k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 4.78k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 5.17k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 5.14k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 4.51k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 5.31k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 3.43k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 4.47k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 4.34k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 4.67k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 3.14k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 4.07k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 3.96k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 4.28k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 3.12k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 4.57k -
EleutherAI/pythia-70m
95.6M • Updated • 333k • 79 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 4.27k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 3.66k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 3.81k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 3.29k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 3.54k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 3.27k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 4.39k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 4.34k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 3.27k -
EleutherAI/pythia-160m
Text Generation • Updated • 2.48M • 39 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 5.12k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 5.41k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 5.35k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 4.18k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 4.58k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 3.25k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 4.4k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 3.23k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 3.39k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 185k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 4.07k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 3.87k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 4.01k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 3.28k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 3.35k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 3.21k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 3.26k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 3.23k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 3.22k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 3.73kNote 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 • 3.66kNote 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 • 3.82kNote 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 • 2.41kNote 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 • 2.39kNote 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 • 2.39kNote 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