lllyasviel's picture
Upload LlamaForCausalLM
564ffe6 verified
|
raw
history blame
155 Bytes
metadata
tags:
  - pytorch
  - trl
  - sft
inference: false

omost-dolphin-2.9-llama3-8b is Omost's llama3-8b model with dolphin-2.9 instruct pretraining in fp16.