[2024-02-01 14:20:07,768] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)
[2024-02-01 14:20:09,368] [WARNING] [runner.py:202:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only.
[2024-02-01 14:20:09,369] [INFO] [runner.py:568:main] cmd = /home/lirenhao/anaconda3/envs/llama_factory/bin/python -u -m deepspeed.launcher.launch --world_info=eyJsb2NhbGhvc3QiOiBbMCwgMSwgMiwgM119 --master_addr=127.0.0.1 --master_port=2345 --enable_each_rank_log=None /home/lirenhao/projects/LLaMA-Factory/src/train_bash.py --deepspeed ds_config.json --stage sft --model_name_or_path /home/lirenhao/pretrained_models/internlm2-chat-7b/ --do_train --dataset cpsycoun --template intern2 --finetuning_type full --lora_target wqkv --output_dir /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9 --overwrite_cache --overwrite_output_dir --per_device_train_batch_size 4 --gradient_accumulation_steps 28 --lr_scheduler_type cosine --logging_steps 10 --save_steps 21 --learning_rate 1e-6 --num_train_epochs 9.0 --plot_loss --fp16
[2024-02-01 14:20:12,819] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)
[2024-02-01 14:20:14,435] [INFO] [launch.py:145:main] WORLD INFO DICT: {'localhost': [0, 1, 2, 3]}
[2024-02-01 14:20:14,436] [INFO] [launch.py:151:main] nnodes=1, num_local_procs=4, node_rank=0
[2024-02-01 14:20:14,436] [INFO] [launch.py:162:main] global_rank_mapping=defaultdict(<class 'list'>, {'localhost': [0, 1, 2, 3]})
[2024-02-01 14:20:14,436] [INFO] [launch.py:163:main] dist_world_size=4
[2024-02-01 14:20:14,436] [INFO] [launch.py:165:main] Setting CUDA_VISIBLE_DEVICES=0,1,2,3
[2024-02-01 14:20:19,797] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)
[2024-02-01 14:20:20,069] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)
[2024-02-01 14:20:20,128] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)
[2024-02-01 14:20:20,157] [INFO] [real_accelerator.py:191:get_accelerator] Setting ds_accelerator to cuda (auto detect)
[2024-02-01 14:20:22,839] [INFO] [comm.py:637:init_distributed] cdb=None
[2024-02-01 14:20:23,347] [INFO] [comm.py:637:init_distributed] cdb=None
[2024-02-01 14:20:23,364] [INFO] [comm.py:637:init_distributed] cdb=None
[2024-02-01 14:20:23,375] [INFO] [comm.py:637:init_distributed] cdb=None
[2024-02-01 14:20:23,376] [INFO] [comm.py:668:init_distributed] Initializing TorchBackend in DeepSpeed with backend nccl
02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
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=None,
ddp_timeout=1800,
debug=[],
deepspeed=ds_config.json,
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=True,
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=28,
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=1e-06,
length_column_name=length,
load_best_model_at_end=False,
local_rank=2,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/runs/Feb01_14-20-22_siat-a100-4-02,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_kwargs={},
lr_scheduler_type=cosine,
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=9.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
overwrite_output_dir=True,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
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=[],
resume_from_checkpoint=None,
run_name=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
save_on_each_node=False,
save_only_model=False,
save_safetensors=True,
save_steps=21,
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,
)
02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Process rank: 0, device: cuda:0, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
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=None,
ddp_timeout=1800,
debug=[],
deepspeed=ds_config.json,
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=True,
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=28,
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=1e-06,
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=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/runs/Feb01_14-20-23_siat-a100-4-02,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_kwargs={},
lr_scheduler_type=cosine,
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=9.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
overwrite_output_dir=True,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
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=[],
resume_from_checkpoint=None,
run_name=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
save_on_each_node=False,
save_only_model=False,
save_safetensors=True,
save_steps=21,
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,
)
[INFO|tokenization_utils_base.py:2025] 2024-02-01 14:20:24,513 >> loading file ./tokenizer.model
[INFO|tokenization_utils_base.py:2025] 2024-02-01 14:20:24,513 >> loading file added_tokens.json
[INFO|tokenization_utils_base.py:2025] 2024-02-01 14:20:24,513 >> loading file special_tokens_map.json
[INFO|tokenization_utils_base.py:2025] 2024-02-01 14:20:24,513 >> loading file tokenizer_config.json
[INFO|tokenization_utils_base.py:2025] 2024-02-01 14:20:24,513 >> loading file tokenizer.json
[INFO|configuration_utils.py:727] 2024-02-01 14:20:24,850 >> loading configuration file /home/lirenhao/pretrained_models/internlm2-chat-7b/config.json
[INFO|configuration_utils.py:727] 2024-02-01 14:20:24,852 >> loading configuration file /home/lirenhao/pretrained_models/internlm2-chat-7b/config.json
[INFO|configuration_utils.py:792] 2024-02-01 14:20:24,854 >> Model config InternLM2Config {
  "_name_or_path": "/home/lirenhao/pretrained_models/internlm2-chat-7b/",
  "architectures": [
    "InternLM2ForCausalLM"
  ],
  "attn_implementation": "eager",
  "auto_map": {
    "AutoConfig": "configuration_internlm2.InternLM2Config",
    "AutoModel": "modeling_internlm2.InternLM2ForCausalLM",
    "AutoModelForCausalLM": "modeling_internlm2.InternLM2ForCausalLM"
  },
  "bias": false,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 14336,
  "max_position_embeddings": 32768,
  "model_type": "internlm2",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 8,
  "pad_token_id": 2,
  "rms_norm_eps": 1e-05,
  "rope_scaling": {
    "factor": 2.0,
    "type": "dynamic"
  },
  "rope_theta": 1000000,
  "tie_word_embeddings": false,
  "torch_dtype": "float16",
  "transformers_version": "4.37.1",
  "use_cache": true,
  "vocab_size": 92544
}

[INFO|modeling_utils.py:3475] 2024-02-01 14:20:24,903 >> loading weights file /home/lirenhao/pretrained_models/internlm2-chat-7b/pytorch_model.bin.index.json
[INFO|modeling_utils.py:1428] 2024-02-01 14:20:24,903 >> Instantiating InternLM2ForCausalLM model under default dtype torch.float16.
[INFO|configuration_utils.py:826] 2024-02-01 14:20:24,905 >> Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "pad_token_id": 2
}

02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Process rank: 1, device: cuda:1, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
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=None,
ddp_timeout=1800,
debug=[],
deepspeed=ds_config.json,
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=True,
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=28,
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=1e-06,
length_column_name=length,
load_best_model_at_end=False,
local_rank=1,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/runs/Feb01_14-20-23_siat-a100-4-02,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_kwargs={},
lr_scheduler_type=cosine,
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=9.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
overwrite_output_dir=True,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
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=[],
resume_from_checkpoint=None,
run_name=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
save_on_each_node=False,
save_only_model=False,
save_safetensors=True,
save_steps=21,
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,
)
02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1
  distributed training: True, compute dtype: torch.float16
02/01/2024 14:20:24 - INFO - llmtuner.hparams.parser - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
auto_find_batch_size=False,
bf16=False,
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=None,
ddp_timeout=1800,
debug=[],
deepspeed=ds_config.json,
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=True,
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=28,
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=1e-06,
length_column_name=length,
load_best_model_at_end=False,
local_rank=3,
log_level=passive,
log_level_replica=warning,
log_on_each_node=True,
logging_dir=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/runs/Feb01_14-20-23_siat-a100-4-02,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_kwargs={},
lr_scheduler_type=cosine,
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=9.0,
optim=adamw_torch,
optim_args=None,
output_dir=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
overwrite_output_dir=True,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=4,
predict_with_generate=False,
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=[],
resume_from_checkpoint=None,
run_name=/home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9,
save_on_each_node=False,
save_only_model=False,
save_safetensors=True,
save_steps=21,
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,
)

Loading checkpoint shards:   0%|          | 0/8 [00:00<?, ?it/s]/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
  return self.fget.__get__(instance, owner)()

Loading checkpoint shards:  12%|█▎        | 1/8 [00:00<00:03,  2.33it/s]
Loading checkpoint shards:  25%|██▌       | 2/8 [00:00<00:02,  2.20it/s]
Loading checkpoint shards:   0%|          | 0/8 [00:00<?, ?it/s]/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
  return self.fget.__get__(instance, owner)()

Loading checkpoint shards:  38%|███▊      | 3/8 [00:01<00:01,  2.74it/s]
Loading checkpoint shards:  50%|█████     | 4/8 [00:01<00:01,  3.09it/s]
Loading checkpoint shards:  12%|█▎        | 1/8 [00:00<00:02,  2.62it/s]
Loading checkpoint shards:   0%|          | 0/8 [00:00<?, ?it/s]/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
  return self.fget.__get__(instance, owner)()

Loading checkpoint shards:   0%|          | 0/8 [00:00<?, ?it/s]/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly.  To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
  return self.fget.__get__(instance, owner)()

Loading checkpoint shards:  62%|██████▎   | 5/8 [00:01<00:00,  3.30it/s]
Loading checkpoint shards:  25%|██▌       | 2/8 [00:00<00:01,  3.18it/s]
Loading checkpoint shards:  12%|█▎        | 1/8 [00:00<00:01,  3.71it/s]
Loading checkpoint shards:  12%|█▎        | 1/8 [00:00<00:01,  4.09it/s]
Loading checkpoint shards:  75%|███████▌  | 6/8 [00:01<00:00,  3.45it/s]
Loading checkpoint shards:  25%|██▌       | 2/8 [00:00<00:01,  4.62it/s]
Loading checkpoint shards:  38%|███▊      | 3/8 [00:00<00:01,  3.42it/s]
Loading checkpoint shards:  25%|██▌       | 2/8 [00:00<00:01,  3.69it/s]
Loading checkpoint shards:  38%|███▊      | 3/8 [00:00<00:01,  4.86it/s]
Loading checkpoint shards:  88%|████████▊ | 7/8 [00:02<00:00,  3.56it/s]
Loading checkpoint shards:  50%|█████     | 4/8 [00:01<00:01,  3.53it/s]
Loading checkpoint shards:  38%|███▊      | 3/8 [00:00<00:01,  3.68it/s]
Loading checkpoint shards:  50%|█████     | 4/8 [00:00<00:00,  4.94it/s]
Loading checkpoint shards: 100%|██████████| 8/8 [00:02<00:00,  3.63it/s]
Loading checkpoint shards: 100%|██████████| 8/8 [00:02<00:00,  3.23it/s]
02/01/2024 14:20:35 - INFO - llmtuner.model.patcher - Gradient checkpointing enabled.
02/01/2024 14:20:35 - INFO - llmtuner.model.adapter - Fine-tuning method: Full

Loading checkpoint shards:  62%|██████▎   | 5/8 [00:01<00:00,  3.51it/s]
Loading checkpoint shards:  50%|█████     | 4/8 [00:01<00:01,  3.58it/s]
Loading checkpoint shards:  62%|██████▎   | 5/8 [00:01<00:00,  4.71it/s]
Loading checkpoint shards:  75%|███████▌  | 6/8 [00:01<00:00,  4.47it/s]
Loading checkpoint shards:  75%|███████▌  | 6/8 [00:01<00:00,  3.51it/s]
Loading checkpoint shards:  62%|██████▎   | 5/8 [00:01<00:00,  3.53it/s]
Loading checkpoint shards:  88%|████████▊ | 7/8 [00:01<00:00,  4.44it/s]
Loading checkpoint shards:  88%|████████▊ | 7/8 [00:02<00:00,  3.54it/s]
Loading checkpoint shards:  75%|███████▌  | 6/8 [00:01<00:00,  3.52it/s]
Loading checkpoint shards: 100%|██████████| 8/8 [00:01<00:00,  4.37it/s]
Loading checkpoint shards: 100%|██████████| 8/8 [00:01<00:00,  4.51it/s]
02/01/2024 14:20:36 - INFO - llmtuner.model.patcher - Gradient checkpointing enabled.
02/01/2024 14:20:36 - INFO - llmtuner.model.adapter - Fine-tuning method: Full

Loading checkpoint shards: 100%|██████████| 8/8 [00:02<00:00,  3.48it/s]
Loading checkpoint shards: 100%|██████████| 8/8 [00:02<00:00,  3.43it/s]
[INFO|modeling_utils.py:4352] 2024-02-01 14:20:36,242 >> All model checkpoint weights were used when initializing InternLM2ForCausalLM.

[INFO|modeling_utils.py:4360] 2024-02-01 14:20:36,242 >> All the weights of InternLM2ForCausalLM were initialized from the model checkpoint at /home/lirenhao/pretrained_models/internlm2-chat-7b/.
If your task is similar to the task the model of the checkpoint was trained on, you can already use InternLM2ForCausalLM for predictions without further training.
[INFO|configuration_utils.py:779] 2024-02-01 14:20:36,247 >> loading configuration file /home/lirenhao/pretrained_models/internlm2-chat-7b/generation_config.json
[INFO|configuration_utils.py:826] 2024-02-01 14:20:36,248 >> Generate config GenerationConfig {
  "bos_token_id": 1,
  "eos_token_id": 2,
  "pad_token_id": 2
}

02/01/2024 14:20:36 - INFO - llmtuner.model.patcher - Gradient checkpointing enabled.
02/01/2024 14:20:36 - INFO - llmtuner.model.adapter - Fine-tuning method: Full

Loading checkpoint shards:  88%|████████▊ | 7/8 [00:01<00:00,  3.43it/s]
Loading checkpoint shards: 100%|██████████| 8/8 [00:02<00:00,  3.42it/s]
Loading checkpoint shards: 100%|██████████| 8/8 [00:02<00:00,  3.50it/s]
02/01/2024 14:20:36 - INFO - llmtuner.model.patcher - Gradient checkpointing enabled.
02/01/2024 14:20:36 - INFO - llmtuner.model.adapter - Fine-tuning method: Full
02/01/2024 14:20:47 - INFO - llmtuner.model.loader - trainable params: 7737708544 || all params: 7737708544 || trainable%: 100.0000
02/01/2024 14:20:48 - INFO - llmtuner.data.template - Add <|im_end|> to stop words.
02/01/2024 14:20:48 - INFO - llmtuner.model.loader - trainable params: 7737708544 || all params: 7737708544 || trainable%: 100.0000
02/01/2024 14:20:48 - INFO - llmtuner.model.loader - trainable params: 7737708544 || all params: 7737708544 || trainable%: 100.0000
02/01/2024 14:20:49 - INFO - llmtuner.data.template - Add <|im_end|> to stop words.
02/01/2024 14:20:49 - WARNING - llmtuner.data.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
02/01/2024 14:20:49 - INFO - llmtuner.data.template - Add <|im_end|> to stop words.
02/01/2024 14:20:49 - INFO - llmtuner.model.loader - trainable params: 7737708544 || all params: 7737708544 || trainable%: 100.0000
02/01/2024 14:20:49 - INFO - llmtuner.data.template - Add <|im_end|> to stop words.
Using custom data configuration default-7bf826ddf73c2f44
Loading Dataset Infos from /home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/datasets/packaged_modules/json
Overwrite dataset info from restored data version if exists.
Loading Dataset info from /home/lirenhao/.cache/huggingface/datasets/json/default-7bf826ddf73c2f44/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96
Found cached dataset json (/home/lirenhao/.cache/huggingface/datasets/json/default-7bf826ddf73c2f44/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96)
Loading Dataset info from /home/lirenhao/.cache/huggingface/datasets/json/default-7bf826ddf73c2f44/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96

Converting format of dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]Caching processed dataset at /home/lirenhao/.cache/huggingface/datasets/json/default-7bf826ddf73c2f44/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96/cache-39a93cb3fc4854d1.arrow

Converting format of dataset:  32%|███▏      | 1000/3134 [00:00<00:00, 7874.85 examples/s]
Converting format of dataset:  64%|██████▍   | 2000/3134 [00:00<00:00, 7860.46 examples/s]
Converting format of dataset:  96%|█████████▌| 3000/3134 [00:00<00:00, 7920.51 examples/s]
Converting format of dataset: 100%|██████████| 3134/3134 [00:00<00:00, 7665.44 examples/s]
02/01/2024 14:20:53 - WARNING - llmtuner.data.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
02/01/2024 14:20:53 - WARNING - llmtuner.data.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.
02/01/2024 14:20:53 - WARNING - llmtuner.data.utils - Checksum failed: missing SHA-1 hash value in dataset_info.json.

Running tokenizer on dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]
Converting format of dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]
Converting format of dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]
Converting format of dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]
Converting format of dataset:  32%|███▏      | 1000/3134 [00:00<00:00, 7894.79 examples/s]
Converting format of dataset:  32%|███▏      | 1000/3134 [00:00<00:00, 7843.15 examples/s]
Converting format of dataset:  32%|███▏      | 1000/3134 [00:00<00:00, 7785.44 examples/s]
Converting format of dataset:  64%|██████▍   | 2000/3134 [00:00<00:00, 7851.76 examples/s]
Converting format of dataset:  64%|██████▍   | 2000/3134 [00:00<00:00, 7817.50 examples/s]
Converting format of dataset:  64%|██████▍   | 2000/3134 [00:00<00:00, 7774.58 examples/s]
Converting format of dataset:  96%|█████████▌| 3000/3134 [00:00<00:00, 7919.08 examples/s]
Converting format of dataset: 100%|██████████| 3134/3134 [00:00<00:00, 7689.85 examples/s]

Converting format of dataset:  96%|█████████▌| 3000/3134 [00:00<00:00, 7876.60 examples/s]
Converting format of dataset:  96%|█████████▌| 3000/3134 [00:00<00:00, 7866.53 examples/s]
Converting format of dataset: 100%|██████████| 3134/3134 [00:00<00:00, 7659.10 examples/s]

Converting format of dataset: 100%|██████████| 3134/3134 [00:00<00:00, 7636.98 examples/s]
Caching processed dataset at /home/lirenhao/.cache/huggingface/datasets/json/default-7bf826ddf73c2f44/0.0.0/8bb11242116d547c741b2e8a1f18598ffdd40a1d4f2a2872c7a28b697434bc96/cache-7cecb244118aac13.arrow

Running tokenizer on dataset:  32%|███▏      | 1000/3134 [00:06<00:13, 161.84 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 2000/3134 [00:10<00:05, 190.86 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 3000/3134 [00:14<00:00, 229.56 examples/s]
Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:14<00:00, 233.78 examples/s]
Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:14<00:00, 215.75 examples/s]
input_ids:
[1, 92543, 9081, 364, 2770, 657, 589, 15358, 17993, 6843, 963, 505, 4576, 11146, 451, 60628, 60384, 60721, 62442, 60752, 4452, 285, 4576, 11146, 451, 60628, 60384, 60721, 62442, 60752, 313, 505, 395, 7659, 1813, 4287, 1762, 560, 505, 8020, 684, 36956, 15358, 31288, 451, 68589, 76659, 71581, 699, 1226, 505, 6342, 442, 517, 11100, 328, 10894, 328, 454, 51978, 756, 285, 4576, 11146, 451, 60628, 60384, 60721, 62442, 60752, 313, 777, 3696, 454, 19187, 19829, 4563, 435, 410, 4287, 12032, 684, 410, 1341, 1893, 569, 6519, 454, 262, 69093, 281, 92542, 364, 92543, 1008, 364, 85064, 60703, 60353, 68856, 68306, 61860, 62703, 69516, 68765, 68984, 60362, 60353, 60376, 68678, 60427, 69944, 60355, 92542, 364, 92543, 525, 11353, 364, 73406, 68865, 68364, 69377, 60353, 86839, 70004, 68364, 69516, 69461, 71677, 68287, 60353, 69029, 68831, 68287, 60355, 68931, 69702, 75326, 71838, 60403, 61860, 62703, 77797, 68540, 60355, 364, 92543, 1008, 364, 74820, 68399, 69088, 60677, 68540, 60353, 61032, 71155, 69059, 60355, 92542, 364, 92543, 525, 11353, 364, 91781, 60353, 73161, 80540, 60415, 82098, 60355, 72010, 71404, 60353, 60403, 76153, 68912, 60381, 74112, 61076, 60504, 364, 92543, 1008, 364, 68856, 68306, 68912, 68326, 75848, 68595, 60353, 69972, 71645, 68473, 68585, 60353, 60404, 68965, 61716, 60418, 68273, 60353, 70124, 70698, 60363, 60355, 68389, 60363, 69667, 68306, 68303, 60353, 61214, 68310, 68758, 68261, 70623, 60355, 92542, 364, 92543, 525, 11353, 364, 72010, 69030, 71711, 61076, 60504, 68522, 60353, 86004, 71645, 68629, 68804, 68592, 69095, 60504, 364, 92543, 1008, 364, 84386, 68268, 68315, 75835, 79506, 60353, 60404, 68965, 72245, 68306, 69377, 60355, 60363, 69836, 60427, 70681, 60353, 69281, 91457, 71102, 62600, 62792, 60425, 60355, 92542, 364, 92543, 525, 11353, 364, 82967, 69068, 87160, 68261, 60504, 68319, 60353, 60403, 70868, 69962, 60871, 69893, 60366, 73603, 68261, 60504, 364, 92543, 1008, 364, 68678, 70219, 92396, 84863, 73603, 68252, 71869, 76758, 60353, 68252, 78650, 68306, 71645, 60355, 60363, 70802, 68626, 71010, 73382, 69893, 60353, 60499, 69361, 61032, 68678, 72415, 60355, 92542, 364, 92543, 525, 11353, 364, 72010, 82120, 68381, 72415, 70860, 69209, 61076, 60504, 68522, 60353, 73880, 60359, 75493, 60359, 72415, 60504, 364, 92543, 1008, 364, 74212, 60353, 60363, 73408, 69836, 73880, 60381, 72415, 60355, 69097, 60353, 88720, 60382, 71343, 68254, 70861, 68892, 60459, 71356, 60586, 60355, 92542, 364, 92543, 525, 11353, 364, 68374, 69209, 73175, 68364, 77514, 61076, 60504, 68522, 60353, 68364, 70033, 60359, 69441, 60359, 68273, 60504, 364, 92543, 1008, 364, 88554, 60355, 60363, 68848, 70033, 81269, 60353, 68965, 60520, 80959, 60355, 68389, 60363, 79837, 80665, 60353, 70465, 70802, 70133, 60355, 92542, 364, 92543, 525, 11353, 364, 76273, 68374, 60353, 69060, 71958, 60353, 68364, 69516, 70848, 69715, 60354, 60355, 81425, 68831, 68637, 60353, 80698, 74131, 73382, 79309, 60355, 364, 92543, 1008, 364, 68369, 61076, 60504, 60462, 69735, 91900, 60827, 60504, 92542, 364, 92543, 525, 11353, 364, 68400, 60353, 69897, 68505, 68364, 69209, 60353, 68908, 69116, 60381, 82567, 69290, 60355, 68265, 60353, 73161, 72826, 68288, 69418, 68304, 68747, 60353, 69068, 60381, 71645, 60359, 68303, 82409, 68615, 69715, 60355, 364, 92543, 1008, 364, 82700, 69460, 70417, 60355, 86492, 60577, 69353, 68301, 60827, 60504, 92542, 364, 92543, 525, 11353, 364, 60577, 68505, 68364, 69209, 68301, 60355, 75630, 82261, 68300, 60353, 68848, 68427, 69836, 73880, 68319, 72415, 68269, 60353, 69095, 68540, 74465, 60504, 2]
inputs:
 <s> <|im_start|>  system
You are an AI assistant whose name is InternLM (书生·浦语).
- InternLM (书生·浦语) is a conversational language model that is developed by Shanghai AI Laboratory (上海人工智能实验室). It is designed to be helpful, honest, and harmless.
- InternLM (书生·浦语) can understand and communicate fluently in the language chosen by the user such as English and 中文.<|im_end|>
 <|im_start|>  user
心理咨询师，我觉得我的胸闷症状越来越严重了，这让我很害怕。<|im_end|>
 <|im_start|> assistant
我能理解你的感受，首先我们要明确你的症状并不是生理问题，而是心理问题。我们可以尝试找出引发你胸闷的心理原因。
 <|im_start|>  user
可是我一直都在找原因，却找不到答案。<|im_end|>
 <|im_start|> assistant
不要着急，我们会一步一步地解决这个问题。你能告诉我，你生活中的压力和困扰吗？
 <|im_start|>  user
我觉得我的压力主要来自于家庭，我和丈夫关系不好，他总是忙于工作，很少关心我。而且我担心我的孩子，怕他们出了什么意外。<|im_end|>
 <|im_start|> assistant
你能详细说说吗？比如，你和丈夫之间的问题具体是什么？
 <|im_start|>  user
我们经常因为一些小事争吵，他总是忽略我的感受。我感到很孤独，就像被困在一个牢笼里。<|im_end|>
 <|im_start|> assistant
这种感觉让你想起了什么？或者，你觉得自己在这段婚姻中失去了什么？
 <|im_start|>  user
让我想想……我觉得我失去了一个温馨的家，一个关爱我的丈夫。我一直在努力维持这段婚姻，但现实却让我失望。<|im_end|>
 <|im_start|> assistant
你能体会到这种失望带来的情绪吗？比如，伤心、愤怒、失望？
 <|im_start|>  user
是的，我经常会感到伤心和失望。有时候，我甚至会怀疑自己的人生是不是选错了路。<|im_end|>
 <|im_start|> assistant
这些情绪会影响你的日常生活吗？比如，你的睡眠、饮食、工作？
 <|im_start|>  user
肯定的。我最近睡眠很差，总是做噩梦。而且我吃得也不好，体重一直在下降。<|im_end|>
 <|im_start|> assistant
了解到这些，我想告诉你，你的症状是可以改善的。我们可以通过心理治疗，帮助你走出这段困境。
 <|im_start|>  user
真的吗？那我要如何做呢？<|im_end|>
 <|im_start|> assistant
首先，我们要了解你的情绪，学会面对和接纳它们。然后，我们会教你怎么表达自己的需求，让你和丈夫、孩子之间的关系得到改善。
 <|im_start|>  user
听起来很有道理。那我们从哪里开始呢？<|im_end|>
 <|im_start|>  assistant
从了解你的情绪开始。试着回想一下，最近一次感到伤心或者失望的时候，是什么原因导致的？</s>
label_ids:
[-100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 73406, 68865, 68364, 69377, 60353, 86839, 70004, 68364, 69516, 69461, 71677, 68287, 60353, 69029, 68831, 68287, 60355, 68931, 69702, 75326, 71838, 60403, 61860, 62703, 77797, 68540, 60355, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 91781, 60353, 73161, 80540, 60415, 82098, 60355, 72010, 71404, 60353, 60403, 76153, 68912, 60381, 74112, 61076, 60504, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 72010, 69030, 71711, 61076, 60504, 68522, 60353, 86004, 71645, 68629, 68804, 68592, 69095, 60504, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 82967, 69068, 87160, 68261, 60504, 68319, 60353, 60403, 70868, 69962, 60871, 69893, 60366, 73603, 68261, 60504, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 72010, 82120, 68381, 72415, 70860, 69209, 61076, 60504, 68522, 60353, 73880, 60359, 75493, 60359, 72415, 60504, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 68374, 69209, 73175, 68364, 77514, 61076, 60504, 68522, 60353, 68364, 70033, 60359, 69441, 60359, 68273, 60504, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 76273, 68374, 60353, 69060, 71958, 60353, 68364, 69516, 70848, 69715, 60354, 60355, 81425, 68831, 68637, 60353, 80698, 74131, 73382, 79309, 60355, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 68400, 60353, 69897, 68505, 68364, 69209, 60353, 68908, 69116, 60381, 82567, 69290, 60355, 68265, 60353, 73161, 72826, 68288, 69418, 68304, 68747, 60353, 69068, 60381, 71645, 60359, 68303, 82409, 68615, 69715, 60355, 2, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, -100, 60577, 68505, 68364, 69209, 68301, 60355, 75630, 82261, 68300, 60353, 68848, 68427, 69836, 73880, 68319, 72415, 68269, 60353, 69095, 68540, 74465, 60504, 2]
labels:
 我能理解你的感受，首先我们要明确你的症状并不是生理问题，而是心理问题。我们可以尝试找出引发你胸闷的心理原因。</s> 不要着急，我们会一步一步地解决这个问题。你能告诉我，你生活中的压力和困扰吗？</s> 你能详细说说吗？比如，你和丈夫之间的问题具体是什么？</s> 这种感觉让你想起了什么？或者，你觉得自己在这段婚姻中失去了什么？</s> 你能体会到这种失望带来的情绪吗？比如，伤心、愤怒、失望？</s> 这些情绪会影响你的日常生活吗？比如，你的睡眠、饮食、工作？</s> 了解到这些，我想告诉你，你的症状是可以改善的。我们可以通过心理治疗，帮助你走出这段困境。</s> 首先，我们要了解你的情绪，学会面对和接纳它们。然后，我们会教你怎么表达自己的需求，让你和丈夫、孩子之间的关系得到改善。</s> 从了解你的情绪开始。试着回想一下，最近一次感到伤心或者失望的时候，是什么原因导致的？</s>
[INFO|training_args.py:1828] 2024-02-01 14:21:08,098 >> PyTorch: setting up devices

Running tokenizer on dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]
Running tokenizer on dataset:   0%|          | 0/3134 [00:00<?, ? examples/s]/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/transformers/training_args.py:1741: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
Detected kernel version 5.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.
[INFO|trainer.py:571] 2024-02-01 14:21:08,153 >> Using auto half precision backend
[2024-02-01 14:21:08,351] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed info: version=0.13.1, git-hash=unknown, git-branch=unknown

Running tokenizer on dataset:  32%|███▏      | 1000/3134 [00:06<00:13, 162.30 examples/s]
Running tokenizer on dataset:  32%|███▏      | 1000/3134 [00:06<00:13, 161.85 examples/s]
Running tokenizer on dataset:  32%|███▏      | 1000/3134 [00:06<00:13, 161.58 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 2000/3134 [00:11<00:06, 172.50 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 2000/3134 [00:12<00:07, 160.87 examples/s]
Running tokenizer on dataset:  64%|██████▍   | 2000/3134 [00:12<00:07, 160.52 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 3000/3134 [00:15<00:00, 213.06 examples/s]
Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:15<00:00, 217.65 examples/s]
Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:15<00:00, 201.27 examples/s]
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/transformers/training_args.py:1741: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(

Running tokenizer on dataset:  96%|█████████▌| 3000/3134 [00:18<00:00, 159.71 examples/s]
Running tokenizer on dataset:  96%|█████████▌| 3000/3134 [00:18<00:00, 159.56 examples/s]
Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:19<00:00, 159.07 examples/s]
Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:19<00:00, 159.20 examples/s]
Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:19<00:00, 159.55 examples/s]

Running tokenizer on dataset: 100%|██████████| 3134/3134 [00:19<00:00, 159.49 examples/s]
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/transformers/training_args.py:1741: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/transformers/training_args.py:1741: FutureWarning: `--push_to_hub_token` is deprecated and will be removed in version 5 of 🤗 Transformers. Use `--hub_token` instead.
  warnings.warn(
[2024-02-01 14:21:41,776] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Flops Profiler Enabled: False
[2024-02-01 14:21:41,778] [INFO] [logging.py:96:log_dist] [Rank 0] Using client Optimizer as basic optimizer
[2024-02-01 14:21:41,778] [INFO] [logging.py:96:log_dist] [Rank 0] Removing param_group that has no 'params' in the basic Optimizer
[2024-02-01 14:21:41,794] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Basic Optimizer = AdamW
[2024-02-01 14:21:41,794] [INFO] [utils.py:56:is_zero_supported_optimizer] Checking ZeRO support for optimizer=AdamW type=<class 'torch.optim.adamw.AdamW'>
[2024-02-01 14:21:41,794] [INFO] [logging.py:96:log_dist] [Rank 0] Creating torch.float16 ZeRO stage 2 optimizer
[2024-02-01 14:21:41,795] [INFO] [stage_1_and_2.py:143:__init__] Reduce bucket size 500000000
[2024-02-01 14:21:41,795] [INFO] [stage_1_and_2.py:144:__init__] Allgather bucket size 500000000
[2024-02-01 14:21:41,795] [INFO] [stage_1_and_2.py:145:__init__] CPU Offload: False
[2024-02-01 14:21:41,795] [INFO] [stage_1_and_2.py:146:__init__] Round robin gradient partitioning: False
[2024-02-01 14:22:01,253] [INFO] [utils.py:791:see_memory_usage] Before initializing optimizer states
[2024-02-01 14:22:01,254] [INFO] [utils.py:792:see_memory_usage] MA 22.12 GB         Max_MA 25.72 GB         CA 25.85 GB         Max_CA 26 GB 
[2024-02-01 14:22:01,254] [INFO] [utils.py:799:see_memory_usage] CPU Virtual Memory:  used = 119.45 GB, percent = 12.4%
[2024-02-01 14:22:01,614] [INFO] [utils.py:791:see_memory_usage] After initializing optimizer states
[2024-02-01 14:22:01,615] [INFO] [utils.py:792:see_memory_usage] MA 36.53 GB         Max_MA 50.95 GB         CA 54.68 GB         Max_CA 55 GB 
[2024-02-01 14:22:01,615] [INFO] [utils.py:799:see_memory_usage] CPU Virtual Memory:  used = 109.91 GB, percent = 11.4%
[2024-02-01 14:22:01,615] [INFO] [stage_1_and_2.py:533:__init__] optimizer state initialized
[2024-02-01 14:22:01,876] [INFO] [utils.py:791:see_memory_usage] After initializing ZeRO optimizer
[2024-02-01 14:22:01,877] [INFO] [utils.py:792:see_memory_usage] MA 36.53 GB         Max_MA 36.53 GB         CA 54.68 GB         Max_CA 55 GB 
[2024-02-01 14:22:01,878] [INFO] [utils.py:799:see_memory_usage] CPU Virtual Memory:  used = 101.72 GB, percent = 10.5%
[2024-02-01 14:22:01,881] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed Final Optimizer = AdamW
[2024-02-01 14:22:01,881] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed using client LR scheduler
[2024-02-01 14:22:01,881] [INFO] [logging.py:96:log_dist] [Rank 0] DeepSpeed LR Scheduler = None
[2024-02-01 14:22:01,881] [INFO] [logging.py:96:log_dist] [Rank 0] step=0, skipped=0, lr=[1e-06], mom=[(0.9, 0.999)]
[2024-02-01 14:22:01,883] [INFO] [config.py:984:print] DeepSpeedEngine configuration:
[2024-02-01 14:22:01,883] [INFO] [config.py:988:print]   activation_checkpointing_config  {
    "partition_activations": false, 
    "contiguous_memory_optimization": false, 
    "cpu_checkpointing": false, 
    "number_checkpoints": null, 
    "synchronize_checkpoint_boundary": false, 
    "profile": false
}
[2024-02-01 14:22:01,883] [INFO] [config.py:988:print]   aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True}
[2024-02-01 14:22:01,883] [INFO] [config.py:988:print]   amp_enabled .................. False
[2024-02-01 14:22:01,883] [INFO] [config.py:988:print]   amp_params ................... False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   autotuning_config ............ {
    "enabled": false, 
    "start_step": null, 
    "end_step": null, 
    "metric_path": null, 
    "arg_mappings": null, 
    "metric": "throughput", 
    "model_info": null, 
    "results_dir": "autotuning_results", 
    "exps_dir": "autotuning_exps", 
    "overwrite": true, 
    "fast": true, 
    "start_profile_step": 3, 
    "end_profile_step": 5, 
    "tuner_type": "gridsearch", 
    "tuner_early_stopping": 5, 
    "tuner_num_trials": 50, 
    "model_info_path": null, 
    "mp_size": 1, 
    "max_train_batch_size": null, 
    "min_train_batch_size": 1, 
    "max_train_micro_batch_size_per_gpu": 1.024000e+03, 
    "min_train_micro_batch_size_per_gpu": 1, 
    "num_tuning_micro_batch_sizes": 3
}
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   bfloat16_enabled ............. False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   checkpoint_parallel_write_pipeline  False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   checkpoint_tag_validation_enabled  True
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   checkpoint_tag_validation_fail  False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   comms_config ................. <deepspeed.comm.config.DeepSpeedCommsConfig object at 0x7f7f6152d840>
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   communication_data_type ...... None
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   compression_config ........... {'weight_quantization': {'shared_parameters': {'enabled': False, 'quantizer_kernel': False, 'schedule_offset': 0, 'quantize_groups': 1, 'quantize_verbose': False, 'quantization_type': 'symmetric', 'quantize_weight_in_forward': False, 'rounding': 'nearest', 'fp16_mixed_quantize': False, 'quantize_change_ratio': 0.001}, 'different_groups': {}}, 'activation_quantization': {'shared_parameters': {'enabled': False, 'quantization_type': 'symmetric', 'range_calibration': 'dynamic', 'schedule_offset': 1000}, 'different_groups': {}}, 'sparse_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'row_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'head_pruning': {'shared_parameters': {'enabled': False, 'method': 'topk', 'schedule_offset': 1000}, 'different_groups': {}}, 'channel_pruning': {'shared_parameters': {'enabled': False, 'method': 'l1', 'schedule_offset': 1000}, 'different_groups': {}}, 'layer_reduction': {'enabled': False}}
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   curriculum_enabled_legacy .... False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   curriculum_params_legacy ..... False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   data_efficiency_config ....... {'enabled': False, 'seed': 1234, 'data_sampling': {'enabled': False, 'num_epochs': 1000, 'num_workers': 0, 'curriculum_learning': {'enabled': False}}, 'data_routing': {'enabled': False, 'random_ltd': {'enabled': False, 'layer_token_lr_schedule': {'enabled': False}}}}
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   data_efficiency_enabled ...... False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   dataloader_drop_last ......... False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   disable_allgather ............ False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   dump_state ................... False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   dynamic_loss_scale_args ...... {'init_scale': 65536, 'scale_window': 1000, 'delayed_shift': 2, 'consecutive_hysteresis': False, 'min_scale': 1}
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_enabled ........... False
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_gas_boundary_resolution  1
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_layer_name ........ bert.encoder.layer
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_layer_num ......... 0
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_max_iter .......... 100
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_stability ......... 1e-06
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_tol ............... 0.01
[2024-02-01 14:22:01,884] [INFO] [config.py:988:print]   eigenvalue_verbose ........... False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   elasticity_enabled ........... False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   flops_profiler_config ........ {
    "enabled": false, 
    "recompute_fwd_factor": 0.0, 
    "profile_step": 1, 
    "module_depth": -1, 
    "top_modules": 1, 
    "detailed": true, 
    "output_file": null
}
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   fp16_auto_cast ............... False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   fp16_enabled ................. True
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   fp16_master_weights_and_gradients  False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   global_rank .................. 0
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   grad_accum_dtype ............. None
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   gradient_accumulation_steps .. 28
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   gradient_clipping ............ 1.0
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   gradient_predivide_factor .... 1.0
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   graph_harvesting ............. False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   hybrid_engine ................ enabled=False max_out_tokens=512 inference_tp_size=1 release_inference_cache=False pin_parameters=True tp_gather_partition_size=8
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   initial_dynamic_scale ........ 65536
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   load_universal_checkpoint .... False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   loss_scale ................... 0
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   memory_breakdown ............. False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   mics_hierarchial_params_gather  False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   mics_shard_size .............. -1
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   monitor_config ............... tensorboard=TensorBoardConfig(enabled=False, output_path='', job_name='DeepSpeedJobName') wandb=WandbConfig(enabled=False, group=None, team=None, project='deepspeed') csv_monitor=CSVConfig(enabled=False, output_path='', job_name='DeepSpeedJobName') enabled=False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   nebula_config ................ {
    "enabled": false, 
    "persistent_storage_path": null, 
    "persistent_time_interval": 100, 
    "num_of_version_in_retention": 2, 
    "enable_nebula_load": true, 
    "load_path": null
}
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   optimizer_legacy_fusion ...... False
[2024-02-01 14:22:01,885] [INFO] [config.py:988:print]   optimizer_name ............... None
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   optimizer_params ............. None
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0, 'pipe_partitioned': True, 'grad_partitioned': True}
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   pld_enabled .................. False
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   pld_params ................... False
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   prescale_gradients ........... False
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   scheduler_name ............... None
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   scheduler_params ............. None
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   seq_parallel_communication_data_type  torch.float32
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   sparse_attention ............. None
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   sparse_gradients_enabled ..... False
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   steps_per_print .............. inf
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   train_batch_size ............. 448
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   train_micro_batch_size_per_gpu  4
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   use_data_before_expert_parallel_  False
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   use_node_local_storage ....... False
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   wall_clock_breakdown ......... False
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   weight_quantization_config ... None
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   world_size ................... 4
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   zero_allow_untested_optimizer  True
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   zero_config .................. stage=2 contiguous_gradients=True reduce_scatter=True reduce_bucket_size=500000000 use_multi_rank_bucket_allreduce=True allgather_partitions=True allgather_bucket_size=500000000 overlap_comm=False load_from_fp32_weights=True elastic_checkpoint=False offload_param=None offload_optimizer=None sub_group_size=1,000,000,000 cpu_offload_param=None cpu_offload_use_pin_memory=None cpu_offload=None prefetch_bucket_size=50,000,000 param_persistence_threshold=100,000 model_persistence_threshold=sys.maxsize max_live_parameters=1,000,000,000 max_reuse_distance=1,000,000,000 gather_16bit_weights_on_model_save=False stage3_gather_fp16_weights_on_model_save=False ignore_unused_parameters=True legacy_stage1=False round_robin_gradients=False zero_hpz_partition_size=1 zero_quantized_weights=False zero_quantized_nontrainable_weights=False zero_quantized_gradients=False mics_shard_size=-1 mics_hierarchical_params_gather=False memory_efficient_linear=True pipeline_loading_checkpoint=False override_module_apply=True
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   zero_enabled ................. True
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   zero_force_ds_cpu_optimizer .. True
[2024-02-01 14:22:01,886] [INFO] [config.py:988:print]   zero_optimization_stage ...... 2
[2024-02-01 14:22:01,887] [INFO] [config.py:974:print_user_config]   json = {
    "train_batch_size": 448, 
    "train_micro_batch_size_per_gpu": 4, 
    "gradient_accumulation_steps": 28, 
    "gradient_clipping": 1.0, 
    "zero_allow_untested_optimizer": true, 
    "fp16": {
        "enabled": true, 
        "loss_scale": 0, 
        "initial_scale_power": 16, 
        "loss_scale_window": 1000, 
        "hysteresis": 2, 
        "min_loss_scale": 1
    }, 
    "zero_optimization": {
        "stage": 2, 
        "allgather_partitions": true, 
        "allgather_bucket_size": 5.000000e+08, 
        "reduce_scatter": true, 
        "reduce_bucket_size": 5.000000e+08, 
        "overlap_comm": false, 
        "contiguous_gradients": true
    }, 
    "steps_per_print": inf, 
    "bf16": {
        "enabled": false
    }
}
[INFO|trainer.py:1721] 2024-02-01 14:22:01,887 >> ***** Running training *****
[INFO|trainer.py:1722] 2024-02-01 14:22:01,887 >>   Num examples = 3,134
[INFO|trainer.py:1723] 2024-02-01 14:22:01,887 >>   Num Epochs = 9
[INFO|trainer.py:1724] 2024-02-01 14:22:01,887 >>   Instantaneous batch size per device = 4
[INFO|trainer.py:1727] 2024-02-01 14:22:01,887 >>   Total train batch size (w. parallel, distributed & accumulation) = 448
[INFO|trainer.py:1728] 2024-02-01 14:22:01,887 >>   Gradient Accumulation steps = 28
[INFO|trainer.py:1729] 2024-02-01 14:22:01,887 >>   Total optimization steps = 63
[INFO|trainer.py:1730] 2024-02-01 14:22:01,889 >>   Number of trainable parameters = 7,737,708,544

  0%|          | 0/63 [00:00<?, ?it/s]/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/deepspeed/runtime/zero/stage_1_and_2.py:1968: UserWarning: The torch.cuda.*DtypeTensor constructors are no longer recommended. It's best to use methods such as torch.tensor(data, dtype=*, device='cuda') to create tensors. (Triggered internally at ../torch/csrc/tensor/python_tensor.cpp:83.)
  overflow_gpu = get_accelerator().ByteTensor([overflow])
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/deepspeed/runtime/zero/stage_1_and_2.py:1968: UserWarning: The torch.cuda.*DtypeTensor constructors are no longer recommended. It's best to use methods such as torch.tensor(data, dtype=*, device='cuda') to create tensors. (Triggered internally at ../torch/csrc/tensor/python_tensor.cpp:83.)
  overflow_gpu = get_accelerator().ByteTensor([overflow])
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/deepspeed/runtime/zero/stage_1_and_2.py:1968: UserWarning: The torch.cuda.*DtypeTensor constructors are no longer recommended. It's best to use methods such as torch.tensor(data, dtype=*, device='cuda') to create tensors. (Triggered internally at ../torch/csrc/tensor/python_tensor.cpp:83.)
  overflow_gpu = get_accelerator().ByteTensor([overflow])
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/deepspeed/runtime/zero/stage_1_and_2.py:1968: UserWarning: The torch.cuda.*DtypeTensor constructors are no longer recommended. It's best to use methods such as torch.tensor(data, dtype=*, device='cuda') to create tensors. (Triggered internally at ../torch/csrc/tensor/python_tensor.cpp:83.)
  overflow_gpu = get_accelerator().ByteTensor([overflow])

  2%|▏         | 1/63 [00:44<45:32, 44.06s/it]
  3%|▎         | 2/63 [01:23<42:01, 41.33s/it]
  5%|▍         | 3/63 [02:04<41:04, 41.08s/it]
  6%|▋         | 4/63 [02:43<39:37, 40.30s/it]
  8%|▊         | 5/63 [03:22<38:23, 39.72s/it][2024-02-01 14:26:04,941] [INFO] [loss_scaler.py:190:update_scale] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536, but hysteresis is 2. Reducing hysteresis to 1

 10%|▉         | 6/63 [04:03<38:08, 40.15s/it][2024-02-01 14:26:44,502] [INFO] [loss_scaler.py:183:update_scale] [deepspeed] OVERFLOW! Rank 0 Skipping step. Attempted loss scale: 65536, reducing to 32768

 11%|█         | 7/63 [04:42<37:17, 39.96s/it]
 13%|█▎        | 8/63 [05:21<36:23, 39.71s/it]
 14%|█▍        | 9/63 [06:01<35:39, 39.62s/it]
 16%|█▌        | 10/63 [06:41<35:16, 39.93s/it]
                                               
{'loss': 2.1176, 'learning_rate': 9.607381059352038e-07, 'epoch': 1.43}

 16%|█▌        | 10/63 [06:41<35:16, 39.93s/it]
 17%|█▋        | 11/63 [07:20<34:22, 39.65s/it]
 19%|█▉        | 12/63 [08:00<33:47, 39.76s/it]
 21%|██        | 13/63 [08:39<32:56, 39.53s/it]
 22%|██▏       | 14/63 [09:20<32:32, 39.85s/it]
 24%|██▍       | 15/63 [09:59<31:45, 39.69s/it]
 25%|██▌       | 16/63 [10:38<30:47, 39.31s/it]
 27%|██▋       | 17/63 [11:19<30:31, 39.82s/it]
 29%|██▊       | 18/63 [11:58<29:51, 39.81s/it]
 30%|███       | 19/63 [12:39<29:15, 39.89s/it]
 32%|███▏      | 20/63 [13:19<28:42, 40.06s/it]
                                               
{'loss': 1.5678, 'learning_rate': 8.117449009293668e-07, 'epoch': 2.86}

 32%|███▏      | 20/63 [13:19<28:42, 40.06s/it]
 33%|███▎      | 21/63 [13:59<27:59, 39.99s/it][INFO|trainer.py:2926] 2024-02-01 14:36:12,897 >> Saving model checkpoint to /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21
[INFO|configuration_utils.py:473] 2024-02-01 14:36:12,902 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/config.json
[INFO|configuration_utils.py:594] 2024-02-01 14:36:12,903 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/generation_config.json
[INFO|modeling_utils.py:2503] 2024-02-01 14:36:40,422 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 4 checkpoint shards. You can find where each parameters has been saved in the index located at /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/model.safetensors.index.json.
[INFO|tokenization_utils_base.py:2433] 2024-02-01 14:36:40,424 >> tokenizer config file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/tokenizer_config.json
[INFO|tokenization_utils_base.py:2442] 2024-02-01 14:36:40,424 >> Special tokens file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/special_tokens_map.json
[2024-02-01 14:36:41,670] [INFO] [logging.py:96:log_dist] [Rank 0] [Torch] Checkpoint global_step21 is about to be saved!
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/nn/modules/module.py:1879: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/nn/modules/module.py:1879: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/nn/modules/module.py:1879: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/nn/modules/module.py:1879: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
[2024-02-01 14:36:41,683] [INFO] [logging.py:96:log_dist] [Rank 0] Saving model checkpoint: /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/global_step21/mp_rank_00_model_states.pt
[2024-02-01 14:36:41,684] [INFO] [torch_checkpoint_engine.py:21:save] [Torch] Saving /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/global_step21/mp_rank_00_model_states.pt...
[2024-02-01 14:37:17,058] [INFO] [torch_checkpoint_engine.py:23:save] [Torch] Saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/global_step21/mp_rank_00_model_states.pt.
[2024-02-01 14:37:17,061] [INFO] [torch_checkpoint_engine.py:21:save] [Torch] Saving /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/global_step21/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2024-02-01 14:38:15,362] [INFO] [torch_checkpoint_engine.py:23:save] [Torch] Saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/global_step21/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2024-02-01 14:38:15,363] [INFO] [engine.py:3477:_save_zero_checkpoint] zero checkpoint saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-21/global_step21/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2024-02-01 14:38:15,363] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step21 is ready now!
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
  warnings.warn(

 35%|███▍      | 22/63 [16:52<54:35, 79.88s/it]
 37%|███▋      | 23/63 [17:32<45:23, 68.08s/it]
 38%|███▊      | 24/63 [18:15<39:17, 60.44s/it]
 40%|███▉      | 25/63 [18:54<34:12, 54.02s/it]
 41%|████▏     | 26/63 [19:33<30:33, 49.54s/it]
 43%|████▎     | 27/63 [20:12<27:49, 46.38s/it]
 44%|████▍     | 28/63 [20:51<25:45, 44.17s/it]
 46%|████▌     | 29/63 [21:31<24:19, 42.92s/it]
 48%|████▊     | 30/63 [22:11<23:07, 42.06s/it]
                                               
{'loss': 1.4173, 'learning_rate': 5.868240888334652e-07, 'epoch': 4.29}

 48%|████▊     | 30/63 [22:11<23:07, 42.06s/it]
 49%|████▉     | 31/63 [22:52<22:17, 41.80s/it]
 51%|█████     | 32/63 [23:32<21:11, 41.02s/it]
 52%|█████▏    | 33/63 [24:10<20:06, 40.20s/it]
 54%|█████▍    | 34/63 [24:49<19:18, 39.96s/it]
 56%|█████▌    | 35/63 [25:30<18:43, 40.11s/it]
 57%|█████▋    | 36/63 [26:10<18:03, 40.13s/it]
 59%|█████▊    | 37/63 [26:49<17:12, 39.70s/it]
 60%|██████    | 38/63 [27:29<16:36, 39.88s/it]
 62%|██████▏   | 39/63 [28:08<15:48, 39.51s/it]
 63%|██████▎   | 40/63 [28:46<15:04, 39.34s/it]
                                               
{'loss': 1.342, 'learning_rate': 3.4075667487415785e-07, 'epoch': 5.71}

 63%|██████▎   | 40/63 [28:46<15:04, 39.34s/it]
 65%|██████▌   | 41/63 [29:27<14:36, 39.84s/it]
 67%|██████▋   | 42/63 [30:07<13:57, 39.87s/it][INFO|trainer.py:2926] 2024-02-01 14:52:21,426 >> Saving model checkpoint to /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42
[INFO|configuration_utils.py:473] 2024-02-01 14:52:21,431 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/config.json
[INFO|configuration_utils.py:594] 2024-02-01 14:52:21,432 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/generation_config.json
[INFO|modeling_utils.py:2503] 2024-02-01 14:52:48,702 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 4 checkpoint shards. You can find where each parameters has been saved in the index located at /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/model.safetensors.index.json.
[INFO|tokenization_utils_base.py:2433] 2024-02-01 14:52:48,704 >> tokenizer config file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/tokenizer_config.json
[INFO|tokenization_utils_base.py:2442] 2024-02-01 14:52:48,704 >> Special tokens file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/special_tokens_map.json
[2024-02-01 14:52:49,843] [INFO] [logging.py:96:log_dist] [Rank 0] [Torch] Checkpoint global_step42 is about to be saved!
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/nn/modules/module.py:1879: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
[2024-02-01 14:52:49,856] [INFO] [logging.py:96:log_dist] [Rank 0] Saving model checkpoint: /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/global_step42/mp_rank_00_model_states.pt
[2024-02-01 14:52:49,856] [INFO] [torch_checkpoint_engine.py:21:save] [Torch] Saving /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/global_step42/mp_rank_00_model_states.pt...
[2024-02-01 14:53:25,041] [INFO] [torch_checkpoint_engine.py:23:save] [Torch] Saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/global_step42/mp_rank_00_model_states.pt.
[2024-02-01 14:53:25,044] [INFO] [torch_checkpoint_engine.py:21:save] [Torch] Saving /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/global_step42/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2024-02-01 14:54:24,364] [INFO] [torch_checkpoint_engine.py:23:save] [Torch] Saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/global_step42/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2024-02-01 14:54:24,364] [INFO] [engine.py:3477:_save_zero_checkpoint] zero checkpoint saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-42/global_step42/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2024-02-01 14:54:24,364] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step42 is ready now!
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
  warnings.warn(

 68%|██████▊   | 43/63 [33:01<26:37, 79.86s/it]
 70%|██████▉   | 44/63 [33:41<21:31, 67.96s/it]
 71%|███████▏  | 45/63 [34:20<17:47, 59.29s/it]
 73%|███████▎  | 46/63 [35:01<15:13, 53.75s/it]
 75%|███████▍  | 47/63 [35:41<13:13, 49.58s/it]
 76%|███████▌  | 48/63 [36:21<11:40, 46.71s/it]
 78%|███████▊  | 49/63 [37:00<10:25, 44.69s/it]
 79%|███████▉  | 50/63 [37:42<09:26, 43.60s/it]
                                               
{'loss': 1.3108, 'learning_rate': 1.3347406408508694e-07, 'epoch': 7.14}

 79%|███████▉  | 50/63 [37:42<09:26, 43.60s/it]
 81%|████████  | 51/63 [38:20<08:24, 42.07s/it]
 83%|████████▎ | 52/63 [39:00<07:34, 41.29s/it]
 84%|████████▍ | 53/63 [39:41<06:52, 41.22s/it]
 86%|████████▌ | 54/63 [40:21<06:07, 40.87s/it]
 87%|████████▋ | 55/63 [41:00<05:22, 40.28s/it]
 89%|████████▉ | 56/63 [41:38<04:39, 39.88s/it]
 90%|█████████ | 57/63 [42:18<03:58, 39.78s/it]
 92%|█████████▏| 58/63 [42:59<03:20, 40.10s/it]
 94%|█████████▎| 59/63 [43:39<02:40, 40.24s/it]
 95%|█████████▌| 60/63 [44:19<02:00, 40.04s/it]
                                               
{'loss': 1.2987, 'learning_rate': 1.5461356885461075e-08, 'epoch': 8.57}

 95%|█████████▌| 60/63 [44:19<02:00, 40.04s/it]
 97%|█████████▋| 61/63 [44:58<01:19, 39.67s/it]
 98%|█████████▊| 62/63 [45:38<00:39, 39.80s/it]
100%|██████████| 63/63 [46:16<00:00, 39.26s/it][INFO|trainer.py:2926] 2024-02-01 15:08:30,323 >> Saving model checkpoint to /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63
[INFO|configuration_utils.py:473] 2024-02-01 15:08:30,328 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/config.json
[INFO|configuration_utils.py:594] 2024-02-01 15:08:30,329 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/generation_config.json
[INFO|modeling_utils.py:2503] 2024-02-01 15:08:57,391 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 4 checkpoint shards. You can find where each parameters has been saved in the index located at /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/model.safetensors.index.json.
[INFO|tokenization_utils_base.py:2433] 2024-02-01 15:08:57,393 >> tokenizer config file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/tokenizer_config.json
[INFO|tokenization_utils_base.py:2442] 2024-02-01 15:08:57,393 >> Special tokens file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/special_tokens_map.json
[2024-02-01 15:08:58,595] [INFO] [logging.py:96:log_dist] [Rank 0] [Torch] Checkpoint global_step63 is about to be saved!
/home/lirenhao/anaconda3/envs/llama_factory/lib/python3.10/site-packages/torch/nn/modules/module.py:1879: UserWarning: Positional args are being deprecated, use kwargs instead. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.nn.Module.state_dict for details.
  warnings.warn(
[2024-02-01 15:08:58,608] [INFO] [logging.py:96:log_dist] [Rank 0] Saving model checkpoint: /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/global_step63/mp_rank_00_model_states.pt
[2024-02-01 15:08:58,608] [INFO] [torch_checkpoint_engine.py:21:save] [Torch] Saving /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/global_step63/mp_rank_00_model_states.pt...
[2024-02-01 15:09:33,948] [INFO] [torch_checkpoint_engine.py:23:save] [Torch] Saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/global_step63/mp_rank_00_model_states.pt.
[2024-02-01 15:09:33,951] [INFO] [torch_checkpoint_engine.py:21:save] [Torch] Saving /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/global_step63/zero_pp_rank_0_mp_rank_00_optim_states.pt...
[2024-02-01 15:10:31,865] [INFO] [torch_checkpoint_engine.py:23:save] [Torch] Saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/global_step63/zero_pp_rank_0_mp_rank_00_optim_states.pt.
[2024-02-01 15:10:31,866] [INFO] [engine.py:3477:_save_zero_checkpoint] zero checkpoint saved /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tmp-checkpoint-63/global_step63/zero_pp_rank_0_mp_rank_00_optim_states.pt
[2024-02-01 15:10:31,866] [INFO] [torch_checkpoint_engine.py:33:commit] [Torch] Checkpoint global_step63 is ready now!
[INFO|trainer.py:1962] 2024-02-01 15:10:32,863 >> 

Training completed. Do not forget to share your model on huggingface.co/models =)



                                               
{'train_runtime': 2910.9748, 'train_samples_per_second': 9.69, 'train_steps_per_second': 0.022, 'train_loss': 1.4981852107577853, 'epoch': 9.0}

100%|██████████| 63/63 [48:30<00:00, 39.26s/it]
100%|██████████| 63/63 [48:30<00:00, 46.21s/it]
[INFO|trainer.py:2926] 2024-02-01 15:10:44,639 >> Saving model checkpoint to /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9
[INFO|configuration_utils.py:473] 2024-02-01 15:10:44,787 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/config.json
[INFO|configuration_utils.py:594] 2024-02-01 15:10:44,788 >> Configuration saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/generation_config.json
[2024-02-01 15:10:49,742] [INFO] [launch.py:347:main] Process 3771596 exits successfully.
[2024-02-01 15:10:49,742] [INFO] [launch.py:347:main] Process 3771597 exits successfully.
[2024-02-01 15:10:49,742] [INFO] [launch.py:347:main] Process 3771598 exits successfully.
[INFO|modeling_utils.py:2503] 2024-02-01 15:11:12,707 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 4 checkpoint shards. You can find where each parameters has been saved in the index located at /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/model.safetensors.index.json.
[INFO|tokenization_utils_base.py:2433] 2024-02-01 15:11:12,709 >> tokenizer config file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/tokenizer_config.json
[INFO|tokenization_utils_base.py:2442] 2024-02-01 15:11:12,709 >> Special tokens file saved in /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/special_tokens_map.json
***** train metrics *****
  epoch                    =        9.0
  train_loss               =     1.4982
  train_runtime            = 0:48:30.97
  train_samples_per_second =       9.69
  train_steps_per_second   =      0.022
Figure saved: /home/lirenhao/projects/LLaMA-Factory/output/9f100e26-d997-46e8-afee-721977a16ca9/training_loss.png
02/01/2024 15:11:14 - WARNING - llmtuner.extras.ploting - No metric eval_loss to plot.
[INFO|modelcard.py:452] 2024-02-01 15:11:14,095 >> Dropping the following result as it does not have all the necessary fields:
{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
[2024-02-01 15:11:17,773] [INFO] [launch.py:347:main] Process 3771595 exits successfully.