Alpha-VLLM / LLaMA2-Accessory

An Open-source Toolkit for LLM Development
https://llama2-accessory.readthedocs.io/
Other
2.72k stars 176 forks source link

The model-type of Multi-Modal Full-Parameter Finetuning #206

Open hekkang opened 1 month ago

hekkang commented 1 month ago

torchrun --nproc-per-node=2 demos/single_turn_mm.py \ --llama_config /path/to/params.json --tokenizer_path /path/to/tokenizer.model \ --pretrained_path /path/to/multimodel_llama

The prompts are as follows: ValueError: Cannot determine llama_type

Then I don't know what type to choose, and there's no meta.json inside the model weight links