Upload llama3_lora_sft.yaml with huggingface_hub

llama3_lora_sft.yaml

@@ -0,0 +1,38 @@
+### model
+model_name_or_path: /root/paddlejob/workspace/env_run/output/output/mistral-7b-inst-v0.3
+
+### method
+stage: sft
+do_train: true
+finetuning_type: full
+
+### dataset
+dataset: alpaca_en_demo_52k
+template: mistral
+cutoff_len: 1024
+max_samples: 10000
+overwrite_cache: true
+preprocessing_num_workers: 16
+
+### output
+output_dir: saves/llama3-8b/lora/sft
+logging_steps: 1
+save_steps: 500
+plot_loss: true
+overwrite_output_dir: true
+
+### train
+per_device_train_batch_size: 1
+gradient_accumulation_steps: 8
+learning_rate: 2.0e-5
+num_train_epochs: 3.0
+lr_scheduler_type: cosine
+warmup_ratio: 0.1
+bf16: true
+ddp_timeout: 180000000
+
+### eval
+val_size: 0.1
+per_device_eval_batch_size: 1
+eval_strategy: steps
+eval_steps: 500