wehub-resource-sync
37 lines
1.1 KiB
Bash
37 lines
1.1 KiB
Bash
# muon: 2 * 65GiB, 4m 14s
|
|
# adam(w): 2 * 78GiB, 1m 19s
|
|
# mcore>=0.16
|
|
PYTORCH_CUDA_ALLOC_CONF='expandable_segments:True' \
|
|
NPROC_PER_NODE=2 \
|
|
CUDA_VISIBLE_DEVICES=0,1 \
|
|
megatron sft \
|
|
--model Qwen/Qwen2.5-7B-Instruct \
|
|
--save_safetensors true \
|
|
--dataset 'AI-ModelScope/alpaca-gpt4-data-zh#500' \
|
|
'AI-ModelScope/alpaca-gpt4-data-en#500' \
|
|
'swift/self-cognition#500' \
|
|
--optimizer dist_muon \
|
|
--tensor_model_parallel_size 2 \
|
|
--sequence_parallel true \
|
|
--micro_batch_size 16 \
|
|
--global_batch_size 16 \
|
|
--recompute_granularity full \
|
|
--recompute_method uniform \
|
|
--recompute_num_layers 1 \
|
|
--finetune true \
|
|
--cross_entropy_loss_fusion true \
|
|
--lr 1e-5 \
|
|
--lr_warmup_fraction 0.05 \
|
|
--min_lr 1e-6 \
|
|
--num_train_epochs 1 \
|
|
--output_dir megatron_output/Qwen2.5-7B-Instruct \
|
|
--save_steps 100 \
|
|
--max_length 2048 \
|
|
--system 'You are a helpful assistant.' \
|
|
--dataloader_num_workers 4 \
|
|
--no_save_optim true \
|
|
--no_save_rng true \
|
|
--dataset_num_proc 4 \
|
|
--model_author swift \
|
|
--model_name swift-robot
|