PolyPythias
- Preview • Updated • 60
EleutherAI/pile-preshuffled-seeds
Updated • 210 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 20.4k • 29 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 4.84k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 4.34k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 4.28k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 2.89k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 3.16k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 3.72k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 4.52k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 2.94k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 4.44k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 1.31k • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 2.83k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 2.73k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 3.08k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 1.55k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 2.5k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 2.38k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 2.7k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 1.54k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 2.97k -
EleutherAI/pythia-70m
95.6M • Updated • 239k • 79 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 2.65k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 2.03k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 2.21k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 1.7k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 1.95k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 1.67k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 2.76k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 2.72k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 1.68k -
EleutherAI/pythia-160m
Text Generation • Updated • 2.45M • 39 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 3.56k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 3.06k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 3.44k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 2.56k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 2.98k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 1.65k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 2.79k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 1.62k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 1.79k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 133k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 2.42k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 2.26k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 2.4k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 1.67k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 1.7k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 1.6k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 1.63k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 1.58k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 1.6k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 2.09kNote 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 • 2.04kNote 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 • 2.18kNote 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-weight-seed1
Text Generation • Updated • 797Note 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 • 774Note 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 • 748Note 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