PolyPythias
- Preview • Updated • 71
EleutherAI/pile-preshuffled-seeds
Updated • 210 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 20.5k • 28 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 4.58k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 4.23k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 4.08k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 2.81k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 2.93k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 3.61k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 4.26k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 2.87k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 4.21k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 785 • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 2.64k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 2.62k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 2.88k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 1.46k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 2.31k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 2.25k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 2.5k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 1.46k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 2.81k -
EleutherAI/pythia-70m
95.6M • Updated • 294k • 79 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 2.5k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 1.97k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 2.12k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 1.65k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 1.83k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 1.59k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 2.64k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 2.62k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 1.57k -
EleutherAI/pythia-160m
Text Generation • Updated • 2.56M • 39 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 2.93k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 2.81k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 2.82k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 2.42k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 2.84k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 1.57k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 2.67k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 1.54k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 1.65k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 136k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 2.01k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 2.17k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 2.01k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 1.59k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 1.6k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 1.51k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 1.53k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 1.49k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 1.51k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 1.69kNote 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-seed2
Text Generation • Updated • 1.91kNote 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 • 1.74kNote 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 • 417Note 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 • 369Note 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 • 347Note 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