09/02/2024 09:22:37 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: False 09/02/2024 09:22:37 - INFO - __main__ - Training parameters TrainingArguments( _n_gpu=1, adafactor=False, adam_beta1=0.9, adam_beta2=0.999, adam_epsilon=1e-08, analysis_dataset=bbh, analysis_mode=False, auto_find_batch_size=False, bf16=True, bf16_full_eval=False, data_seed=9, dataloader_drop_last=False, dataloader_num_workers=0, dataloader_persistent_workers=False, dataloader_pin_memory=True, ddp_backend=None, ddp_broadcast_buffers=None, ddp_bucket_cap_mb=None, ddp_find_unused_parameters=None, ddp_timeout=1800, debug=[], deepspeed=None, disable_tqdm=False, dispatch_batches=None, do_eval=False, do_predict=False, do_train=True, eval_accumulation_steps=None, eval_delay=0, eval_steps=None, evaluation_strategy=no, fp16=False, fp16_backend=auto, fp16_full_eval=False, fp16_opt_level=O1, fsdp=[, ], fsdp_config={'fsdp_transformer_layer_cls_to_wrap': ['LlamaDecoderLayer'], 'fsdp_backward_prefetch': 'backward_pre', 'limit_all_gathers': 'true', 'use_orig_params': 'true', 'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False}, fsdp_min_num_params=0, fsdp_transformer_layer_cls_to_wrap=None, full_determinism=False, gradient_accumulation_steps=32, gradient_checkpointing=False, gradient_checkpointing_kwargs=None, greater_is_better=None, group_by_length=False, half_precision_backend=auto, hub_always_push=False, hub_model_id=None, hub_private_repo=False, hub_strategy=every_save, hub_token=, ignore_data_skip=False, include_inputs_for_metrics=False, include_num_input_tokens_seen=False, include_tokens_per_second=False, jit_mode_eval=False, label_names=None, label_smoothing_factor=0.0, learning_rate=2e-05, length_column_name=length, load_best_model_at_end=False, local_rank=0, log_level=passive, log_level_replica=warning, log_on_each_node=True, logging_dir=../out/llama2-7b-p0.05-lora-seed9/runs/Sep02_09-22-37_dlc19oe6tjflhwaw-master-0, logging_first_step=False, logging_nan_inf_filter=True, logging_steps=1.0, logging_strategy=steps, lr_scheduler_kwargs={}, lr_scheduler_type=linear, max_grad_norm=1.0, max_steps=-1, metric_for_best_model=None, mp_parameters=, neftune_noise_alpha=None, no_cuda=False, num_train_epochs=4.0, optim=adamw_torch, optim_args=None, output_dir=../out/llama2-7b-p0.05-lora-seed9, overwrite_output_dir=True, past_index=-1, per_device_eval_batch_size=8, per_device_train_batch_size=1, prediction_loss_only=False, push_to_hub=False, push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=, ray_scope=last, remove_unused_columns=True, report_to=['wandb'], resume_from_checkpoint=None, run_name=../out/llama2-7b-p0.05-lora-seed9, save_on_each_node=False, save_only_model=False, save_safetensors=True, save_steps=500, save_strategy=epoch, save_total_limit=None, seed=0, skip_memory_metrics=True, split_batches=False, tf32=False, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, torchdynamo=None, tpu_metrics_debug=False, tpu_num_cores=None, train_dataset_names=None, use_cpu=False, use_ipex=False, use_legacy_prediction_loop=False, use_mps_device=False, warmup_ratio=0.03, warmup_steps=0, weight_decay=0.0, ) 09/02/2024 09:22:37 - INFO - __main__ - Model parameters ModelArguments(model_name_or_path='/mnt/data_large/ccy/Llama-2-7b-hf', config_name=None, tokenizer_name=None, cache_dir=None, use_fast_tokenizer=True, model_revision='main', use_auth_token=False, torch_dtype=None, lora=True, lora_r=128, lora_alpha=512.0, lora_dropout=0.1, lora_target_modules=['q_proj', 'k_proj', 'v_proj', 'o_proj']) 09/02/2024 09:22:37 - INFO - __main__ - Dataset parameters DataArguments(train_files=['./data/train/processed/flan_v2/flan_v2_data.jsonl', './data/train/processed/cot/cot_data.jsonl', './data/train/processed/dolly/dolly_data.jsonl', './data/train/processed/oasst1/oasst1_data.jsonl'], overwrite_cache=False, preprocessing_num_workers=None, max_seq_length=2048, sample_data_seed=42, percentage=0.05) [INFO|tokenization_utils_base.py:2024] 2024-09-02 09:22:37,635 >> loading file tokenizer.model [INFO|tokenization_utils_base.py:2024] 2024-09-02 09:22:37,635 >> loading file tokenizer.json [INFO|tokenization_utils_base.py:2024] 2024-09-02 09:22:37,635 >> loading file added_tokens.json [INFO|tokenization_utils_base.py:2024] 2024-09-02 09:22:37,635 >> loading file special_tokens_map.json [INFO|tokenization_utils_base.py:2024] 2024-09-02 09:22:37,635 >> loading file tokenizer_config.json Using custom data configuration default-f9134d95c1961a9a 09/02/2024 09:22:38 - INFO - datasets.builder - Using custom data configuration default-f9134d95c1961a9a Loading Dataset Infos from /home/pai/envs/less/lib/python3.10/site-packages/datasets/packaged_modules/json 09/02/2024 09:22:38 - INFO - datasets.info - Loading Dataset Infos from /home/pai/envs/less/lib/python3.10/site-packages/datasets/packaged_modules/json Overwrite dataset info from restored data version if exists. 09/02/2024 09:22:38 - INFO - datasets.builder - Overwrite dataset info from restored data version if exists. Loading Dataset info from /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7 09/02/2024 09:22:38 - INFO - datasets.info - Loading Dataset info from /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7 Found cached dataset json (/root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7) 09/02/2024 09:22:38 - INFO - datasets.builder - Found cached dataset json (/root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7) Loading Dataset info from /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7 09/02/2024 09:22:38 - INFO - datasets.info - Loading Dataset info from /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7 /mnt/data_large/ccy/Llama-2-7b-hf ['./data/train/processed/flan_v2/flan_v2_data.jsonl', './data/train/processed/cot/cot_data.jsonl', './data/train/processed/dolly/dolly_data.jsonl', './data/train/processed/oasst1/oasst1_data.jsonl'] ['dataset', 'id', 'messages'] Process #0 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00000_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #0 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00000_of_00010.arrow Process #1 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00001_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #1 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00001_of_00010.arrow Process #2 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00002_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #2 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00002_of_00010.arrow Process #3 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00003_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #3 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00003_of_00010.arrow Process #4 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00004_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #4 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00004_of_00010.arrow Process #5 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00005_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #5 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00005_of_00010.arrow Process #6 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00006_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #6 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00006_of_00010.arrow Process #7 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00007_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #7 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00007_of_00010.arrow Process #8 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00008_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #8 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00008_of_00010.arrow Process #9 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00009_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Process #9 will write at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_00009_of_00010.arrow Loading cached processed dataset at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_*_of_00010.arrow 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Loading cached processed dataset at /root/.cache/huggingface/datasets/json/default-f9134d95c1961a9a/0.0.0/c8d2d9508a2a2067ab02cd118834ecef34c3700d143b31835ec4235bf10109f7/cache-7ad18c1f20ceb8df_*_of_00010.arrow Concatenating 10 shards 09/02/2024 09:22:38 - INFO - datasets.arrow_dataset - Concatenating 10 shards [INFO|configuration_utils.py:737] 2024-09-02 09:22:38,728 >> loading configuration file /mnt/data_large/ccy/Llama-2-7b-hf/config.json [INFO|configuration_utils.py:802] 2024-09-02 09:22:38,729 >> Model config LlamaConfig { "_name_or_path": "/mnt/data_large/ccy/Llama-2-7b-hf", "architectures": [ "LlamaForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "eos_token_id": 2, "hidden_act": "silu", "hidden_size": 4096, "initializer_range": 0.02, "intermediate_size": 11008, "max_position_embeddings": 4096, "model_type": "llama", "num_attention_heads": 32, "num_hidden_layers": 32, "num_key_value_heads": 32, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 10000.0, "tie_word_embeddings": false, "transformers_version": "4.36.2", "use_cache": true, "vocab_size": 32000 } [INFO|modeling_utils.py:3341] 2024-09-02 09:22:38,764 >> loading weights file /mnt/data_large/ccy/Llama-2-7b-hf/model.safetensors.index.json [INFO|configuration_utils.py:826] 2024-09-02 09:22:38,765 >> Generate config GenerationConfig { "bos_token_id": 1, "eos_token_id": 2 } Loading checkpoint shards: 0%| | 0/2 [00:00