Closed lichenyigit closed 11 months ago
DISC-LawLLM has been training on self-cognition. This behavior may be very hard to override, at least not possible with only ~10 data.
需要多少个数据才会有效果呢?
@Charlie-XIAO 可以抽空回答一下问题吗?
@lichenyigit I have no exact answer for this, but there may need to be at least hundreds.
后来,又进行了10000次训练力度,还是没有效果。请问是哪个步骤出了问题? 数据集: 新闻Q&A.json 训练脚本:
torchrun --nproc_per_node 1 src/train_bash.py \
--stage sft \
--model_name_or_path ShengbinYue/DISC-LawLLM \
--do_train \
--dataset yiqi5-fun \
--template baichuan2 \
--finetuning_type lora \
--lora_rank 8 \
--lora_target W_pack \
--output_dir /home/DISC-output-checkpoint \
--overwrite_output_dir \
--overwrite_cache \
--per_device_train_batch_size 4 \
--per_device_eval_batch_size 4 \
--gradient_accumulation_steps 8 \
--preprocessing_num_workers 16 \
--lr_scheduler_type cosine \
--logging_steps 10 \
--save_steps 100 \
--eval_steps 100 \
--learning_rate 1e-5 \
--max_grad_norm 0.5 \
--num_train_epochs 10000.0 \
--evaluation_strategy steps \
--load_best_model_at_end \
--plot_loss \
--fp16 \
--val_size 0.01
@Charlie-XIAO
Moving to #29, closing this one as completed.
你好我要微调数据。现在用【我是谁】来做测试数据。 微调之后,感觉没有一点效果。 还请看一下什么问题?
数据集如下: 0.json
微调脚本如下: torchrun --nproc_per_node 1 src/train_bash.py \ --stage sft \ --model_name_or_path ShengbinYue/DISC-LawLLM \ --do_train \ --dataset yiqi \ --template baichuan2 \ --finetuning_type lora \ --lora_rank 8 \ --lora_target W_pack \ --output_dir output_checkpoint\ --overwrite_cache \ --per_device_train_batch_size 4 \ --per_device_eval_batch_size 4 \ --gradient_accumulation_steps 8 \ --preprocessing_num_workers 16 \ --lr_scheduler_type cosine \ --logging_steps 10 \ --save_steps 100 \ --eval_steps 100 \ --learning_rate 1e-4 \ --max_grad_norm 0.5 \ --num_train_epochs 3000.0 \ --evaluation_strategy steps \ --load_best_model_at_end \ --plot_loss \ --fp16 \ --val_size 0.01
导出脚本如下: python src/export_model.py \ --model_name_or_path ShengbinYue/DISC-LawLLM \ --template baichuan2 \ --finetuning_type lora \ --checkpoint_dir output_checkpoint \ --export_dir export_model