bf16: true
cutoff_len: 2048
dataset: solidity_contract
dataset_dir: data
ddp_timeout: 180000000
do_train: true
finetuning_type: lora
flash_attn: auto
gradient_accumulation_steps: 2
learning_rate: 5.0e-05
logging_steps: 5
lora_alpha: 32
lora_dropout: 0.1
lora_rank: 16
lora_target: all
lr_scheduler_type: cosine
max_grad_norm: 1.0
max_samples: 100000
model_name_or_path: deepseek-ai/deepseek-coder-6.7b-base
num_train_epochs: 2.0
optim: adamw_torch
output_dir: saves/DeepSeek-Coder-6.7B-Base/lora/solidity_lora_finetuned
packing: false
per_device_train_batch_size: 16
plot_loss: true
preprocessing_num_workers: 16
report_to: none
save_steps: 100
stage: sft
template: default
warmup_steps: 0