ngme-llama-264M / README.md
PatrickHaller's picture
Create README.md
cd96607
metadata
datasets:
  - allenai/c4
language:
  - en
pipeline_tag: text-generation

NGME-LLama 264M

  • Trained on 4 A6000 for ~4 days
  • Trained ~4 Billion (4 * 16 * 768 * 100_000) Tokens
  • On C4 Corpus