hiyouga commited on
Commit
49871be
1 Parent(s): 7d68667

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -17,7 +17,7 @@ inference: false
17
  A bilingual instruction-tuned LoRA model of https://huggingface.co/baichuan-inc/baichuan-7B
18
 
19
  - Instruction-following datasets used: alpaca, alpaca-zh, codealpaca
20
- - Training framework: https://github.com/hiyouga/LLaMA-Efficient-Tuning
21
 
22
  Please follow the [baichuan-7B License](https://huggingface.co/baichuan-inc/baichuan-7B/resolve/main/baichuan-7B%20%E6%A8%A1%E5%9E%8B%E8%AE%B8%E5%8F%AF%E5%8D%8F%E8%AE%AE.pdf) to use this model.
23
 
@@ -42,7 +42,7 @@ inputs = inputs.to("cuda")
42
  generate_ids = model.generate(**inputs, max_new_tokens=256, streamer=streamer)
43
  ```
44
 
45
- You could also alternatively launch a CLI demo by using the script in https://github.com/hiyouga/LLaMA-Efficient-Tuning
46
 
47
  ```bash
48
  python src/cli_demo.py --template default --model_name_or_path hiyouga/baichuan-7b-sft
@@ -50,7 +50,7 @@ python src/cli_demo.py --template default --model_name_or_path hiyouga/baichuan-
50
 
51
  ---
52
 
53
- You could reproduce our results with the following scripts using [LLaMA-Efficient-Tuning](https://github.com/hiyouga/LLaMA-Efficient-Tuning):
54
 
55
  ```bash
56
  CUDA_VISIBLE_DEVICES=0 python src/train_bash.py \
@@ -61,7 +61,7 @@ CUDA_VISIBLE_DEVICES=0 python src/train_bash.py \
61
  --template default \
62
  --finetuning_type lora \
63
  --lora_rank 16 \
64
- --lora_target W_pack,o_proj,gate_proj,down_proj,up_proj \
65
  --output_dir baichuan_lora \
66
  --overwrite_cache \
67
  --per_device_train_batch_size 8 \
 
17
  A bilingual instruction-tuned LoRA model of https://huggingface.co/baichuan-inc/baichuan-7B
18
 
19
  - Instruction-following datasets used: alpaca, alpaca-zh, codealpaca
20
+ - Training framework: https://github.com/hiyouga/LLaMA-Factory
21
 
22
  Please follow the [baichuan-7B License](https://huggingface.co/baichuan-inc/baichuan-7B/resolve/main/baichuan-7B%20%E6%A8%A1%E5%9E%8B%E8%AE%B8%E5%8F%AF%E5%8D%8F%E8%AE%AE.pdf) to use this model.
23
 
 
42
  generate_ids = model.generate(**inputs, max_new_tokens=256, streamer=streamer)
43
  ```
44
 
45
+ You could also alternatively launch a CLI demo by using the script in https://github.com/hiyouga/LLaMA-Factory
46
 
47
  ```bash
48
  python src/cli_demo.py --template default --model_name_or_path hiyouga/baichuan-7b-sft
 
50
 
51
  ---
52
 
53
+ You could reproduce our results with the following scripts using [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory):
54
 
55
  ```bash
56
  CUDA_VISIBLE_DEVICES=0 python src/train_bash.py \
 
61
  --template default \
62
  --finetuning_type lora \
63
  --lora_rank 16 \
64
+ --lora_target all \
65
  --output_dir baichuan_lora \
66
  --overwrite_cache \
67
  --per_device_train_batch_size 8 \