assistant_tag: assistant
bf16: 'True'
content_tag: value
cutoff_len: '16384'
dataset: mlfoundations-dev/stratos_verified_mix
dataset_dir: ONLINE
ddp_timeout: '180000000'
deepspeed: /opt/ml/code/zero3_offload.json
do_train: 'True'
enable_liger_kernel: 'False'
finetuning_type: full
formatting: sharegpt
global_batch_size: '96'
gradient_accumulation_steps: '3'
hub_model_id: mlfoundations-dev/DCFT-Stratos-Verified-114k-7B-4gpus
learning_rate: 1e-05
logging_steps: '1'
lr_scheduler_type: cosine
max_samples: '1000000'
messages: conversations
model_name_or_path: Qwen/Qwen2.5-7B-Instruct
num_train_epochs: '3.0'
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'
report_to: wandb
role_tag: from
run_name: Stratos-35k-32B-DCFT
save_steps: '100'
stage: sft
template: qwen25
user_tag: user
warmup_ratio: '0.1'