-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathqst-70b.sh
43 lines (43 loc) · 1.11 KB
/
qst-70b.sh
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
python qst.py \
--model_name_or_path Llama-2-70b-chat-hf \
--output_dir ./output/QST-70b \
--logging_steps 10 \
--save_strategy steps \
--data_seed 42 \
--save_steps 200 \
--save_total_limit 40 \
--evaluation_strategy steps \
--eval_dataset_size 1024 \
--max_eval_samples 1000 \
--per_device_eval_batch_size 1 \
--max_new_tokens 32 \
--dataloader_num_workers 3 \
--group_by_length \
--logging_strategy steps \
--remove_unused_columns False \
--do_train \
--do_eval \
--do_mmlu_eval \
--lora_r 64 \
--lora_alpha 16 \
--lora_modules all \
--double_quant \
--quant_type nf4 \
--bf16 \
--bits 4 \
--warmup_ratio 0.03 \
--lr_scheduler_type constant \
--gradient_checkpointing False\
--dataset oasst1 \
--source_max_len 512 \
--target_max_len 512 \
--per_device_train_batch_size 1 \
--gradient_accumulation_steps 16 \
--max_steps 1875 \
--eval_steps 187 \
--learning_rate 0.0001 \
--adam_beta2 0.999 \
--max_grad_norm 0.3 \
--lora_dropout 0.05 \
--weight_decay 0.0 \
--seed 0