cutoff_len: 10000
dataset: qwen2_vl_dora
dataset_dir: data
ddp_timeout: 180000000
do_train: true
eval_steps: 30
eval_strategy: steps
finetuning_type: lora
flash_attn: auto
fp16: true
gradient_accumulation_steps: 8
include_num_input_tokens_seen: true
learning_rate: 5.0e-05
logging_steps: 5
lora_alpha: 16
lora_dropout: 0
lora_rank: 8
lora_target: all
lr_scheduler_type: cosine
max_grad_norm: 1.0
max_samples: 100000
model_name_or_path: llava-hf/LLaVA-NeXT-Video-7B-hf
num_train_epochs: 5.0
optim: adamw_torch
output_dir: saves/LLaVA-NeXT-Video-7B-Chat/lora/250212_Abroad_LLaVANeXTVideo_7B
packing: false
per_device_eval_batch_size: 2
per_device_train_batch_size: 2
plot_loss: true
preprocessing_num_workers: 16
report_to: none
save_steps: 30
stage: sft
template: llava_next_video
trust_remote_code: true
val_size: 0.2
warmup_steps: 0