b7e279b200
Model: OhhTuRnz/train-KSPDG-Alpaca-LLaMA-Chat-8B-DoRA-16-8-adamw_8bit_unquantized_10_files Source: Original Platform
62 lines
1.5 KiB
YAML
62 lines
1.5 KiB
YAML
top.adapter_path: []
|
|
top.booster: flashattn2
|
|
top.finetuning_type: lora
|
|
top.model_name: LLaMA3-8B-Chat
|
|
top.quantization_bit: none
|
|
top.rope_scaling: none
|
|
top.template: alpaca
|
|
top.visual_inputs: false
|
|
train.additional_target: ''
|
|
train.badam_mode: layer
|
|
train.badam_switch_interval: 50
|
|
train.badam_switch_mode: ascending
|
|
train.badam_update_ratio: 0.05
|
|
train.batch_size: 2
|
|
train.compute_type: bf16
|
|
train.create_new_adapter: false
|
|
train.cutoff_len: 1024
|
|
train.dataset:
|
|
- KSPDG-LLaMA-3-Alpaca-50
|
|
train.dataset_dir: data
|
|
train.freeze_extra_modules: ''
|
|
train.freeze_trainable_layers: 2
|
|
train.freeze_trainable_modules: all
|
|
train.galore_rank: 16
|
|
train.galore_scale: 0.25
|
|
train.galore_target: all
|
|
train.galore_update_interval: 200
|
|
train.gradient_accumulation_steps: 2
|
|
train.learning_rate: 1e-4
|
|
train.logging_steps: 5
|
|
train.lora_alpha: 8
|
|
train.lora_dropout: 0.05
|
|
train.lora_rank: 16
|
|
train.lora_target: ''
|
|
train.loraplus_lr_ratio: 0
|
|
train.lr_scheduler_type: cosine
|
|
train.max_grad_norm: '1'
|
|
train.max_samples: '100000'
|
|
train.neftune_alpha: 0
|
|
train.num_train_epochs: '3.0'
|
|
train.optim: adamw_8bit
|
|
train.packing: false
|
|
train.ppo_score_norm: false
|
|
train.ppo_whiten_rewards: false
|
|
train.pref_beta: 0.1
|
|
train.pref_ftx: 0
|
|
train.pref_loss: sigmoid
|
|
train.report_to: true
|
|
train.resize_vocab: false
|
|
train.reward_model: null
|
|
train.save_steps: 100
|
|
train.shift_attn: false
|
|
train.training_stage: Supervised Fine-Tuning
|
|
train.upcast_layernorm: false
|
|
train.use_badam: false
|
|
train.use_dora: true
|
|
train.use_galore: false
|
|
train.use_llama_pro: false
|
|
train.use_rslora: false
|
|
train.val_size: 0.01
|
|
train.warmup_steps: 0
|