@@ -148,6 +148,8 @@ bash examples/full_multi_gpu/predict.sh
|
||||
|
||||
#### 合并 LoRA 适配器
|
||||
|
||||
注:请勿使用量化后的模型或 `quantization_bit` 参数来合并 LoRA 适配器。
|
||||
|
||||
```bash
|
||||
CUDA_VISIBLE_DEVICES=0 llamafactory-cli export examples/merge_lora/llama3_lora_sft.yaml
|
||||
```
|
||||
|
||||
Reference in New Issue
Block a user