datablations
/

lm5-2b8-55b-c4

Model card Files Files and versions Metrics Training metrics Community

lm5-2b8-55b-c4 / README.md

Muennighoff's picture

Create README.md

a559ad4 about 2 years ago

|

history blame contribute delete

93 Bytes

GPT model with UL2 objective (Did not perform better than causal language modeling objective)