ros/LLaMA-Factory
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

update readme

Browse Source 
Former-commit-id: 01267eee0da0bffb3f0c0378e2e60d14e05585c4
This commit is contained in:
hiyouga
2023-12-18 15:46:45 +08:00
parent 5e7bb5fe73
commit 74575f8922
2 changed files with 7 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@@ -104,7 +104,7 @@ Compared to ChatGLM's [P-Tuning](https://github.com/THUDM/ChatGLM2-6B/tree/main/
| [LLaMA-2](https://huggingface.co/meta-llama) | 7B/13B/70B | q_proj,v_proj | llama2 |
| [Mistral](https://huggingface.co/mistralai) | 7B | q_proj,v_proj | mistral |
| [Mixtral](https://huggingface.co/mistralai) | 8x7B | q_proj,v_proj | mistral |
| [Phi-1.5](https://huggingface.co/microsoft/phi-1_5) | 1.3B | Wqkv | - |
| [Phi-1.5/2](https://huggingface.co/microsoft) | 1.3B/2.7B | Wqkv | - |
| [Qwen](https://github.com/QwenLM/Qwen) | 1.8B/7B/14B/72B | c_attn | qwen |
| [XVERSE](https://github.com/xverse-ai) | 7B/13B/65B | q_proj,v_proj | xverse |
@@ -126,7 +126,7 @@ Please refer to [constants.py](src/llmtuner/extras/constants.py) for a full list
| DPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
> [!NOTE]
> Use `--quantization_bit 4/8` argument to enable QLoRA.
> Use `--quantization_bit 4` argument to enable QLoRA.
## Provided Datasets
@@ -482,7 +482,7 @@ python src/export_model.py \
> Merging LoRA weights into a quantized model is not supported.
> [!TIP]
> Use `--export_quantization_bit 4` and `--export_quantization_dataset data/c4_demo.json` to quantize the model.
> Use `--export_quantization_bit 4` and `--export_quantization_dataset data/c4_demo.json` to quantize the model after merging the LoRA weights.
### API Demo