|
--- |
|
library_name: peft |
|
license: apache-2.0 |
|
datasets: |
|
- fnlp/moss-003-sft-data |
|
language: |
|
- zh |
|
- en |
|
--- |
|
## Training procedure |
|
|
|
|
|
The following `bitsandbytes` quantization config was used during training: |
|
- load_in_8bit: False |
|
- load_in_4bit: True |
|
- llm_int8_threshold: 6.0 |
|
- llm_int8_skip_modules: None |
|
- llm_int8_enable_fp32_cpu_offload: False |
|
- llm_int8_has_fp16_weight: False |
|
- bnb_4bit_quant_type: nf4 |
|
- bnb_4bit_use_double_quant: True |
|
- bnb_4bit_compute_dtype: float16 |
|
### Framework versions |
|
|
|
|
|
- PEFT 0.4.0.dev0 |
|
|
|
|
|
## 使用方法 |
|
|
|
git clone https://huggingface.co/Laurie/baichuan-7b-qlora-moss |
|
cd baichuan-7b-qlora-moss |
|
python src/web_demo.py \ |
|
--model_name_or_path baichuan-inc/baichuan-7B \ |
|
--checkpoint_dir . |
