| [2023-12-19 17:47:31,804] [INFO] [real_accelerator.py:158:get_accelerator] Setting ds_accelerator to cuda (auto detect) | |
| /root/miniconda3/envs/textgen/lib/python3.10/site-packages/trl/trainer/ppo_config.py:141: UserWarning: The `optimize_cuda_cache` arguement will be deprecated soon, please use `optimize_device_cache` instead. | |
| warnings.warn( | |
| 12/19/2023 17:47:36 - WARNING - llmtuner.model.parser - `ddp_find_unused_parameters` needs to be set as False for LoRA in DDP training. | |
| /root/miniconda3/envs/textgen/lib/python3.10/site-packages/transformers/training_args.py:1751: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead. | |
| warnings.warn( | |
| 12/19/2023 17:47:36 - INFO - llmtuner.model.parser - Process rank: 0, device: cuda:0, n_gpu: 2 | |
| distributed training: True, compute dtype: torch.bfloat16 | |
| 12/19/2023 17:47:36 - INFO - llmtuner.model.parser - Training/evaluation parameters Seq2SeqTrainingArguments( | |
| _n_gpu=2, | |
| adafactor=False, | |
| adam_beta1=0.9, | |
| adam_beta2=0.999, | |
| adam_epsilon=1e-08, | |
| auto_find_batch_size=False, | |
| bf16=True, | |
| bf16_full_eval=False, | |
| data_seed=None, | |
| 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=False, | |
| ddp_timeout=1800, | |
| debug=[], | |
| deepspeed=None, | |
| disable_tqdm=False, | |
| dispatch_batches=None, | |
| do_eval=False, | |
| do_predict=True, | |
| do_train=False, | |
| 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={'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, | |
| generation_config=None, | |
| generation_max_length=None, | |
| generation_num_beams=None, | |
| gradient_accumulation_steps=1, | |
| 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=<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=5e-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=./models/sft/phi-2-sft-alpaca_gpt4_en-1/Predict_20/runs/Dec19_17-47-36_autodl-container-f11a41911a-e496153c, | |
| logging_first_step=False, | |
| logging_nan_inf_filter=True, | |
| logging_steps=500, | |
| 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=3.0, | |
| optim=adamw_torch, | |
| optim_args=None, | |
| output_dir=./models/sft/phi-2-sft-alpaca_gpt4_en-1/Predict_20, | |
| overwrite_output_dir=False, | |
| past_index=-1, | |
| per_device_eval_batch_size=1, | |
| per_device_train_batch_size=8, | |
| predict_with_generate=True, | |
| prediction_loss_only=False, | |
| push_to_hub=False, | |
| push_to_hub_model_id=None, | |
| push_to_hub_organization=None, | |
| push_to_hub_token=<PUSH_TO_HUB_TOKEN>, | |
| ray_scope=last, | |
| remove_unused_columns=True, | |
| report_to=['tensorboard', 'wandb'], | |
| resume_from_checkpoint=None, | |
| run_name=./models/sft/phi-2-sft-alpaca_gpt4_en-1/Predict_20, | |
| save_on_each_node=False, | |
| save_only_model=False, | |
| save_safetensors=True, | |
| save_steps=500, | |
| save_strategy=steps, | |
| save_total_limit=None, | |
| seed=42, | |
| skip_memory_metrics=True, | |
| sortish_sampler=False, | |
| split_batches=False, | |
| tf32=None, | |
| torch_compile=False, | |
| torch_compile_backend=None, | |
| torch_compile_mode=None, | |
| torchdynamo=None, | |
| tpu_metrics_debug=False, | |
| tpu_num_cores=None, | |
| use_cpu=False, | |
| use_ipex=False, | |
| use_legacy_prediction_loop=False, | |
| use_mps_device=False, | |
| warmup_ratio=0.0, | |
| warmup_steps=0, | |
| weight_decay=0.0, | |
| ) | |
| 12/19/2023 17:47:36 - INFO - llmtuner.data.loader - Loading dataset alpaca_gpt4_data_en.json... | |
| [WARNING|logging.py:314] 2023-12-19 17:47:37,929 >> Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. | |
| Loading checkpoint shards: 0%| | 0/2 [00:00<?, ?it/s] Loading checkpoint shards: 50%|█████ | 1/2 [00:00<00:00, 1.75it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.79it/s] Loading checkpoint shards: 100%|██████████| 2/2 [00:00<00:00, 2.56it/s] | |
| 12/19/2023 17:47:38 - INFO - llmtuner.model.adapter - Fine-tuning method: LoRA | |
| 12/19/2023 17:47:39 - INFO - llmtuner.model.adapter - Merged 1 adapter(s). | |
| 12/19/2023 17:47:39 - INFO - llmtuner.model.adapter - Loaded adapter(s): ./models/sft/phi-2-sft-alpaca_gpt4_en-1 | |
| 12/19/2023 17:47:39 - INFO - llmtuner.model.loader - trainable params: 0 || all params: 2779683840 || trainable%: 0.0000 | |
| 12/19/2023 17:47:39 - INFO - llmtuner.model.loader - This IS expected that the trainable params is 0 if you are using model for inference only. | |
| 12/19/2023 17:47:39 - INFO - llmtuner.data.template - Add pad token: <|endoftext|> | |
| [WARNING|logging.py:314] 2023-12-19 17:47:40,715 >> You're using a CodeGenTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding. | |
| /root/miniconda3/envs/textgen/lib/python3.10/site-packages/torch/nn/parallel/_functions.py:68: UserWarning: Was asked to gather along dimension 0, but all input tensors were scalars; will instead unsqueeze and return a vector. | |
| warnings.warn('Was asked to gather along dimension 0, but all ' | |
| input_ids: | |
| [50256, 32, 8537, 1022, 257, 11040, 2836, 290, 281, 11666, 4430, 8796, 13, 383, 8796, 3607, 7613, 11, 6496, 11, 290, 23507, 7429, 284, 262, 2836, 338, 2683, 13, 198, 20490, 25, 13786, 1115, 9040, 329, 10589, 5448, 13, 198, 48902, 25] | |
| inputs: | |
| <|endoftext|>A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. | |
| Human: Give three tips for staying healthy. | |
| Assistant: | |
| 0%| | 0/10 [00:00<?, ?it/s] 20%|██ | 2/10 [00:10<00:43, 5.46s/it] 30%|███ | 3/10 [00:12<00:26, 3.85s/it] 40%|████ | 4/10 [00:20<00:31, 5.22s/it] 50%|█████ | 5/10 [00:23<00:22, 4.47s/it] 60%|██████ | 6/10 [00:26<00:16, 4.15s/it] 70%|███████ | 7/10 [00:39<00:21, 7.02s/it] 80%|████████ | 8/10 [00:46<00:13, 6.96s/it] 90%|█████████ | 9/10 [00:51<00:06, 6.40s/it] 100%|██████████| 10/10 [01:01<00:00, 7.51s/it]Building prefix dict from the default dictionary ... | |
| Loading model from cache /tmp/jieba.cache | |
| Loading model cost 0.578 seconds. | |
| Prefix dict has been built successfully. | |
| 100%|██████████| 10/10 [01:02<00:00, 6.28s/it] | |
| ***** predict metrics ***** | |
| predict_bleu-4 = 49.0534 | |
| predict_rouge-1 = 54.9625 | |
| predict_rouge-2 = 31.0959 | |
| predict_rouge-l = 39.8761 | |
| predict_runtime = 0:01:10.55 | |
| predict_samples_per_second = 0.283 | |
| predict_steps_per_second = 0.142 | |
| 12/19/2023 17:48:51 - INFO - llmtuner.train.sft.trainer - Saving prediction results to ./models/sft/phi-2-sft-alpaca_gpt4_en-1/Predict_20/generated_predictions.jsonl | |