Orca-2-13b-SFT-v4 / README.md
Locutusque's picture
Update README.md
4ec1ddf
|
raw
history blame
568 Bytes
metadata
license: llama2
language:
  - en
datasets:
  - Locutusque/InstructMix

The "meta-llama/Llama-2-13b-hf" model fully fine-tuned on Locutusque/InstructMix. The current weights are the outcome of a test run, having been trained on only 4,000 examples. More epochs are planned for this model. Despite the small training session, the model still achieved a good average loss of about 0.9.

Make sure to comply with the llama2 license. Please read it before using this model.

This model was trained on the following chat template: "<|USER|> message <|ASSISTANT|> message"