PolyPythias
- Preview • Updated • 65
EleutherAI/pile-preshuffled-seeds
Updated • 169 • 1Note Training data information for each seed.
-
EleutherAI/pythia-14m-deduped
Text Generation • 39.2M • Updated • 35.1k • 28 -
EleutherAI/pythia-14m-seed1
Text Generation • Updated • 4.11k -
EleutherAI/pythia-14m-seed2
Text Generation • Updated • 4.18k -
EleutherAI/pythia-14m-seed3
Text Generation • Updated • 3.96k -
EleutherAI/pythia-14m-seed4
Text Generation • Updated • 2.78k -
EleutherAI/pythia-14m-seed5
Text Generation • Updated • 2.81k -
EleutherAI/pythia-14m-seed6
Text Generation • Updated • 3.54k -
EleutherAI/pythia-14m-seed7
Text Generation • Updated • 4.12k -
EleutherAI/pythia-14m-seed8
Text Generation • Updated • 2.82k -
EleutherAI/pythia-14m-seed9
Text Generation • Updated • 4.09k -
EleutherAI/pythia-31m-deduped
Text Generation • 55.7M • Updated • 938 • 5 -
EleutherAI/pythia-31m-seed1
Text Generation • Updated • 2.57k -
EleutherAI/pythia-31m-seed2
Text Generation • Updated • 2.58k -
EleutherAI/pythia-31m-seed3
Text Generation • Updated • 2.8k -
EleutherAI/pythia-31m-seed4
Text Generation • Updated • 1.43k -
EleutherAI/pythia-31m-seed5
Text Generation • Updated • 2.22k -
EleutherAI/pythia-31m-seed6
Text Generation • Updated • 2.2k -
EleutherAI/pythia-31m-seed7
Text Generation • Updated • 2.42k -
EleutherAI/pythia-31m-seed8
Text Generation • Updated • 1.41k -
EleutherAI/pythia-31m-seed9
Text Generation • Updated • 2.73k -
EleutherAI/pythia-70m
95.6M • Updated • 283k • 79 -
EleutherAI/pythia-70m-seed1
Text Generation • Updated • 2.39k -
EleutherAI/pythia-70m-seed2
Text Generation • Updated • 1.87k -
EleutherAI/pythia-70m-seed3
Text Generation • Updated • 2.01k -
EleutherAI/pythia-70m-seed4
Text Generation • Updated • 1.53k -
EleutherAI/pythia-70m-seed5
Text Generation • Updated • 1.71k -
EleutherAI/pythia-70m-seed6
Text Generation • Updated • 1.47k -
EleutherAI/pythia-70m-seed7
Text Generation • Updated • 2.5k -
EleutherAI/pythia-70m-seed8
Text Generation • Updated • 2.5k -
EleutherAI/pythia-70m-seed9
Text Generation • Updated • 1.44k -
EleutherAI/pythia-160m
Text Generation • Updated • 2.6M • 39 -
EleutherAI/pythia-160m-seed1
Text Generation • 0.2B • Updated • 2.85k -
EleutherAI/pythia-160m-seed2
Text Generation • 0.2B • Updated • 2.74k -
EleutherAI/pythia-160m-seed3
Text Generation • 0.2B • Updated • 2.74k -
EleutherAI/pythia-160m-seed4
Text Generation • Updated • 2.35k • 1 -
EleutherAI/pythia-160m-seed5
Text Generation • Updated • 2.75k -
EleutherAI/pythia-160m-seed6
Text Generation • Updated • 1.48k -
EleutherAI/pythia-160m-seed7
Text Generation • Updated • 2.58k -
EleutherAI/pythia-160m-seed8
Text Generation • Updated • 1.44k -
EleutherAI/pythia-160m-seed9
Text Generation • Updated • 1.56k -
EleutherAI/pythia-410m
Text Generation • 0.5B • Updated • 129k • 36 -
EleutherAI/pythia-410m-seed1
Text Generation • Updated • 1.88k -
EleutherAI/pythia-410m-seed2
Text Generation • Updated • 2.1k -
EleutherAI/pythia-410m-seed3
Text Generation • Updated • 1.93k -
EleutherAI/pythia-410m-seed4
Text Generation • Updated • 1.48k -
EleutherAI/pythia-410m-seed5
Text Generation • Updated • 1.46k -
EleutherAI/pythia-410m-seed6
Text Generation • Updated • 1.44k • 1 -
EleutherAI/pythia-410m-seed7
Text Generation • Updated • 1.43k -
EleutherAI/pythia-410m-seed8
Text Generation • Updated • 1.39k -
EleutherAI/pythia-410m-seed9
Text Generation • Updated • 1.44k
EleutherAI/pythia-160m-data-seed1
Text Generation • Updated • 1.64kNote 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 • 1.84kNote 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 • 1.65kNote 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 • 366Note 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 • 320Note 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 • 302Note 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