mlabonne/TwinLlama-3.1-8B-DPO

🤗 Hugging Face sourcetext-generationapache-2.08B params16 GBsafetensorsHF checksums availableupdated today
No torrent yet

Uploaded model

  • Developed by: mlabonne
  • License: apache-2.0
  • Finetuned from model : mlabonne/TwinLlama-3.1-8B

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.