File size: 1,145 Bytes
8d1cda9 |
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 |
adam_beta1: 0.9
adam_beta2: 0.999
assistant_tag: gpt
bf16: true
content_tag: value
cutoff_len: 2048
dataset: /p/data1/mmlaion/marianna/lf_datasets/mlfoundations-dev/stackoverflow_500000_samples
dataset_dir: ONLINE
ddp_timeout: 180000000
deepspeed: dcft/train/zero3.json
do_train: true
enable_liger_kernel: false
eval_strategy: 'no'
finetuning_type: full
formatting: sharegpt
global_batch_size: 512
gradient_accumulation_steps: 4
gradient_checkpointing: true
hub_model_id: mlfoundations-dev/mlfoundations-dev_stackoverflow_500000_samples
learning_rate: 5.0e-06
logging_steps: 10
lr_scheduler_type: constant
max_grad_norm: 1
messages: conversations
model_name_or_path: meta-llama/Meta-Llama-3.1-8B
neat_packing: true
num_train_epochs: 3.0
output_dir: /p/data1/mmlaion/marianna/dcft_checkpoints/mlfoundations-dev_stackoverflow_500000_samples_meta-llama/Meta-Llama-3.1-8B
overwrite_cache: true
overwrite_output_dir: false
packing: true
per_device_train_batch_size: 4
plot_loss: false
preprocessing_num_workers: 16
push_to_db: false
push_to_hub: false
role_tag: from
save_strategy: epoch
stage: sft
template: llama3
user_tag: human
weight_decay: 0.1
|