p12475398
/
LLaMA-Factory-Mirror
forked from p04798526/LLaMA-Factory-Mirror
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

change: change llama2_lora_sft.yaml

This commit is contained in:
wql 2024-08-12 16:51:45 +08:00
parent 1ee249021b
commit 41c42f67a2
1 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
examples/train_lora/llama2_lora_sft.yaml
View File

@ -5,31 +5,31 @@ model_name_or_path: modelscope/Llama-2-7b-ms
stage: sft stage: sft
do_train: true do_train: true
finetuning_type: lora finetuning_type: lora
lora_target: all lora_target: q_proj,v_proj
### dataset ### dataset
dataset: alpaca_zh dataset: alpaca_zh
template: llama2 template: llama2
cutoff_len: 1024 cutoff_len: 1024
max_samples: 1000 max_samples: 100000
overwrite_cache: true overwrite_cache: true
preprocessing_num_workers: 16 preprocessing_num_workers: 16
### output ### output
output_dir: ./saves/LLaMA2-7B/lora/train_24_8_12_15_46 output_dir: ./saves/LLaMA2-7B/lora/train_24_8_12_16_46
logging_steps: 10 logging_steps: 5
save_steps: 500 save_steps: 100
plot_loss: true plot_loss: true
overwrite_output_dir: true overwrite_output_dir: true
### train ### train
per_device_train_batch_size: 2 per_device_train_batch_size: 2
gradient_accumulation_steps: 8 gradient_accumulation_steps: 8
learning_rate: 1.0e-4 learning_rate: 5e-5
num_train_epochs: 3.0 num_train_epochs: 5.0
lr_scheduler_type: cosine lr_scheduler_type: cosine
warmup_ratio: 0.1 warmup_ratio: 0.1
bf16: true fp16: true
ddp_timeout: 180000000 ddp_timeout: 180000000
### eval ### eval