llama-factory/config/llama3_8b_lora_sft.yaml

### model
# model_name_or_path: gradientai/Llama-3-8B-Instruct-Gradient-1048k
model_name_or_path: meta-llama/Meta-Llama-3-8B-Instruct

### method
stage: sft
do_train: true
finetuning_type: lora
lora_target: all
quantization_bit: 4                     # use 4-bit QLoRA
loraplus_lr_ratio: 16.0                 # use LoRA+ with lambda=16.0
# use_unsloth: true                       # use UnslothAI's LoRA optimization for 2x faster training
upcast_layernorm: true
 
### dataset
dataset: alpaca_mac
template: llama3
cutoff_len: 1024
max_samples: 500
overwrite_cache: true
preprocessing_num_workers: 16

### output
# output_dir: saves/llama3-8b/lora/sft
output_dir: /content/llama3-8b/
logging_steps: 10
save_steps: 100
plot_loss: true
overwrite_output_dir: true
# resume_from_checkpoint: true

### train
per_device_train_batch_size: 1
gradient_accumulation_steps: 8
learning_rate: 1.0e-4
num_train_epochs: 6.0
lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: true
ddp_timeout: 180000000

### eval
val_size: 0.01
per_device_eval_batch_size: 1
eval_strategy: steps
eval_steps: 560

report_to: none