PolyPythias
- Preview • Updated • 61
EleutherAI/pile-preshuffled-seeds
Updated • 69 • 1Note Training data information for each seed.
EleutherAI/pythia-14m
Text Generation • Updated • 205k • 21EleutherAI/pythia-14m-seed1
Updated • 16.5kEleutherAI/pythia-14m-seed2
Updated • 1.14kEleutherAI/pythia-14m-seed3
Updated • 965EleutherAI/pythia-14m-seed4
Updated • 912EleutherAI/pythia-14m-seed5
Updated • 829EleutherAI/pythia-14m-seed6
Updated • 513EleutherAI/pythia-14m-seed7
Updated • 512EleutherAI/pythia-14m-seed8
Updated • 511EleutherAI/pythia-14m-seed9
Updated • 507EleutherAI/pythia-31m
Text Generation • Updated • 91.8k • 5EleutherAI/pythia-31m-seed1
Updated • 2.73kEleutherAI/pythia-31m-seed2
Updated • 531EleutherAI/pythia-31m-seed3
Updated • 362EleutherAI/pythia-31m-seed4
Updated • 353EleutherAI/pythia-31m-seed5
Updated • 356EleutherAI/pythia-31m-seed6
Updated • 50EleutherAI/pythia-31m-seed7
Updated • 51EleutherAI/pythia-31m-seed8
Updated • 51EleutherAI/pythia-31m-seed9
Updated • 51EleutherAI/pythia-70m
Updated • 130k • 66EleutherAI/pythia-70m-seed1
Updated • 5.91kEleutherAI/pythia-70m-seed2
Updated • 605EleutherAI/pythia-70m-seed3
Updated • 422EleutherAI/pythia-70m-seed4
Updated • 418EleutherAI/pythia-70m-seed5
Updated • 423EleutherAI/pythia-70m-seed6
Updated • 113EleutherAI/pythia-70m-seed7
Updated • 107EleutherAI/pythia-70m-seed8
Updated • 101EleutherAI/pythia-70m-seed9
Updated • 100EleutherAI/pythia-160m
Text Generation • Updated • 165k • 31EleutherAI/pythia-160m-seed1
Text Generation • Updated • 3kEleutherAI/pythia-160m-seed2
Text Generation • Updated • 1.15kEleutherAI/pythia-160m-seed3
Text Generation • Updated • 986EleutherAI/pythia-160m-seed4
Updated • 318 • 1EleutherAI/pythia-160m-seed5
Updated • 318EleutherAI/pythia-160m-seed6
Updated • 10EleutherAI/pythia-160m-seed7
Updated • 8EleutherAI/pythia-160m-seed8
Updated • 8EleutherAI/pythia-160m-seed9
Updated • 8EleutherAI/pythia-410m
Text Generation • Updated • 82.4k • 24EleutherAI/pythia-410m-seed1
Updated • 1.37kEleutherAI/pythia-410m-seed2
Updated • 3EleutherAI/pythia-410m-seed3
Updated • 3EleutherAI/pythia-410m-seed4
UpdatedEleutherAI/pythia-410m-seed5
UpdatedEleutherAI/pythia-410m-seed6
UpdatedEleutherAI/pythia-410m-seed7
UpdatedEleutherAI/pythia-410m-seed8
UpdatedEleutherAI/pythia-410m-seed9
Updated
EleutherAI/pythia-160m-data-seed1
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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
UpdatedNote 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