Upload gemma2_train_dev_test_qlora_test2.yml

gemma2_train_dev_test_qlora_test2.yml

@@ -0,0 +1,67 @@
+base_model: google/gemma-2-9b-it
+model_type: AutoModelForCausalLM
+tokenizer_type: AutoTokenizer
+
+load_in_8bit: false
+load_in_4bit: true
+strict: false
+
+chat_template: gemma
+datasets:
+  - path: frcp/malpyung_yoyak_train_dev_test
+    type: alpaca
+    chat_template: gemma
+    drop_system_message: true
+val_set_size: 0.0
+output_dir: ./outputs/gemma2_train_dev_test_qlora_test2
+
+adapter: qlora
+lora_r: 64
+lora_alpha: 64
+lora_dropout: 0.05
+lora_target_linear: true
+
+sequence_len: 4096
+sample_packing: false
+eval_sample_packing: false
+pad_to_sequence_len: true
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_name:
+wandb_log_model:
+
+
+gradient_accumulation_steps: 2
+micro_batch_size: 4
+num_epochs: 10
+optimizer: adamw_bnb_8bit
+lr_scheduler: cosine
+learning_rate: 2e-5
+
+train_on_inputs: false
+group_by_length: false
+bf16: auto
+fp16:
+tf32: true
+
+gradient_checkpointing: true
+early_stopping_patience:
+resume_from_checkpoint:
+local_rank:
+logging_steps: 1
+xformers_attention:
+flash_attention: true
+
+warmup_ratio: 0
+evals_per_epoch:
+eval_table_size:
+eval_max_new_tokens: 0
+saves_per_epoch: 0
+debug:
+deepspeed:
+weight_decay: 0.0
+fsdp:
+fsdp_config:
+special_tokens: