Upload configs.yaml with huggingface_hub

configs.yaml

@@ -1,39 +1,37 @@
 assistant_tag: gpt
-bf16: 'True'
+bf16: true
 content_tag: value
-cutoff_len: '16384'
-dataloader_num_workers: '4'
-dataloader_persistent_workers: 'True'
-dataloader_pin_memory: 'True'
-dataset: mlfoundations-dev/fig1_all_s1
+cutoff_len: 16384
+dataloader_num_workers: 4
+dataloader_persistent_workers: true
+dataloader_pin_memory: true
+dataset: mlfoundations-dev/s1
 dataset_dir: ONLINE
-ddp_timeout: '180000000'
-deepspeed: /opt/ml/code/zero3.json
-do_train: 'True'
-enable_liger_kernel: 'True'
+ddp_timeout: 180000000
+deepspeed: dcft/train/zero3.json
+do_train: true
+enable_liger_kernel: true
 finetuning_type: full
-formatting: sharegpt
-global_batch_size: '96'
-gradient_accumulation_steps: '12'
-hub_model_id: mlfoundations-dev/fig1_all_s1
-learning_rate: 2e-05
-logging_steps: '1'
+global_batch_size: 96
+gradient_accumulation_steps: 24
+hub_model_id: mlfoundations-dev/s1
+learning_rate: 2.0e-05
+logging_steps: 1
 lr_scheduler_type: cosine
 messages: conversations
 model_name_or_path: Qwen/Qwen2.5-7B-Instruct
-num_train_epochs: '7.0'
-output_dir: /opt/ml/model
-overwrite_cache: 'True'
-per_device_train_batch_size: '1'
-plot_loss: 'True'
-preprocessing_num_workers: '16'
-push_to_db: 'True'
-push_to_hub: 'True'
+num_train_epochs: 7.0
+output_dir: /data/cat/ws/ryma833h-dcft/checkpoints/s1
+overwrite_cache: true
+per_device_train_batch_size: 1
+plot_loss: true
+preprocessing_num_workers: 16
+push_to_db: true
+push_to_hub: true
 report_to: wandb
 role_tag: from
-run_name: fig1_all_s1
 save_strategy: epoch
 stage: sft
 template: qwen25
 user_tag: human
-warmup_ratio: '0.1'
+warmup_ratio: 0.1