SCPO_V_1-5_7B
This repository contains a LLaVA-LLaMA 7B checkpoint trained with SCPO/DPO-style data.
Model Details
- Architecture:
LlavaLlamaForCausalLM
- Model type:
llava_llama
- Precision:
bfloat16
- Vision tower:
openai/clip-vit-large-patch14-336
- Weights format: sharded
safetensors
Notes
Please verify the license and redistribution terms of the base model and training data before making this repository public.
