README.md

metadata

language:
  - en
tags:
  - pytorch
  - causal-lm
  - pythia
license: apache-2.0
datasets:
  - Anthropic/hh-rlhf

Infos

Pythia-1b supervised finetuned with Anthropic-hh-rlhf dataset for 1 epoch (sft-model), before DPO (paper) with same dataset for 1 epoch.

wandb log

See Pythia-1b for model details (paper).

Benchmark results:

Zero shot

Results for the base model are taken from the Pythia paper.

Five shot