===================================BUG REPORT===================================
Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/usr/local/cuda/extras/CUPTI/lib64'), PosixPath('/usr/local/nvidia/lib')}
warn(msg)
CUDA SETUP: CUDA version lower than 11 are currently not supported for LLM.int8(). You will be only to use 8-bit optimizers and quantization routines!!
CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so
CUDA SETUP: Highest compute capability among GPUs detected: 7.0
CUDA SETUP: Detected CUDA version 102
/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: Compute capability < 7.5 detected! Only slow 8-bit matmul is supported for your GPU!
warn(msg)
CUDA SETUP: Loading binary /home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/bitsandbytes/libbitsandbytes_cuda102_nocublaslt.so...
07/12/2023 16:16:51 - INFO - utils.common - Process rank: -1, device: cuda:0, n_gpu: 8
distributed training: False, 16-bits training: True
07/12/2023 16:16:51 - INFO - utils.common - Training/evaluation parameters Seq2SeqTrainingArguments(
_n_gpu=8,
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_pin_memory=True,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=None,
ddp_timeout=1800,
debug=[],
deepspeed=None,
disable_tqdm=False,
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={'fsdp_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=4,
gradient_checkpointing=False,
greater_is_better=None,
group_by_length=False,
half_precision_backend=auto,
hub_model_id=None,
hub_private_repo=False,
hub_strategy=every_save,
hub_token=,
ignore_data_skip=False,
include_inputs_for_metrics=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=1e-05,
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=path_to_ppo_checkpoint/runs/Jul12_16-16-51_notebook-905609ac-c0e6-4275-af61-d7bde586ccb6,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=1.0,
optim=adamw_torch,
optim_args=None,
output_dir=path_to_ppo_checkpoint,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=2,
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=,
ray_scope=last,
remove_unused_columns=True,
report_to=['tensorboard'],
resume_from_checkpoint=None,
run_name=path_to_ppo_checkpoint,
save_on_each_node=False,
save_safetensors=False,
save_steps=1000,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=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_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
xpu_backend=None,
)
07/12/2023 16:16:51 - INFO - utils.common - Loading dataset alpaca_gpt4_data_en.json...
07/12/2023 16:18:11 - INFO - datasets.builder - Using custom data configuration default-f5c93679fc52bdba
07/12/2023 16:18:11 - INFO - datasets.info - Loading Dataset Infos from /home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/datasets/packaged_modules/json
07/12/2023 16:18:11 - INFO - datasets.builder - Generating dataset json (/home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4)
Downloading and preparing dataset json/default to /home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4...
Downloading data files: 100%|███████████████████| 1/1 [00:00<00:00, 7073.03it/s]
07/12/2023 16:18:11 - INFO - datasets.download.download_manager - Downloading took 0.0 min
07/12/2023 16:18:11 - INFO - datasets.download.download_manager - Checksum Computation took 0.0 min
Extracting data files: 100%|█████████████████████| 1/1 [00:00<00:00, 176.61it/s]
07/12/2023 16:18:11 - INFO - datasets.builder - Generating train split
07/12/2023 16:18:12 - INFO - datasets.utils.info_utils - Unable to verify splits sizes.
Dataset json downloaded and prepared to /home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4. Subsequent calls will reuse this data.
100%|████████████████████████████████████████████| 1/1 [00:00<00:00, 198.98it/s]
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file ice_text.model
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file added_tokens.json
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file special_tokens_map.json
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file tokenizer_config.json
[INFO|configuration_utils.py:666] 2023-07-12 16:18:12,776 >> loading configuration file THUDM/chatglm-6b/config.json
[INFO|configuration_utils.py:666] 2023-07-12 16:18:12,781 >> loading configuration file THUDM/chatglm-6b/config.json
[INFO|configuration_utils.py:720] 2023-07-12 16:18:12,782 >> Model config ChatGLMConfig {
"_name_or_path": "THUDM/chatglm-6b",
"architectures": [
"ChatGLMModel"
],
"auto_map": {
"AutoConfig": "configuration_chatglm.ChatGLMConfig",
"AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration",
"AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration"
},
"bos_token_id": 130004,
"eos_token_id": 130005,
"gmask_token_id": 130001,
"hidden_size": 4096,
"inner_hidden_size": 16384,
"layernorm_epsilon": 1e-05,
"mask_token_id": 130000,
"max_sequence_length": 2048,
"model_type": "chatglm",
"num_attention_heads": 32,
"num_layers": 28,
"pad_token_id": 3,
"position_encoding_2d": true,
"pre_seq_len": null,
"prefix_projection": false,
"quantization_bit": 0,
"torch_dtype": "float16",
"transformers_version": "4.28.0",
"use_cache": true,
"vocab_size": 130528
}
Loading checkpoint shards: 100%|██████████████████| 8/8 [00:13<00:00, 1.65s/it]
[INFO|modeling_utils.py:3190] 2023-07-12 16:18:26,160 >> All model checkpoint weights were used when initializing ChatGLMForConditionalGeneration.
[INFO|modeling_utils.py:3199] 2023-07-12 16:18:26,160 >> All the weights of ChatGLMForConditionalGeneration were initialized from the model checkpoint at THUDM/chatglm-6b.
If your task is similar to the task the model of the checkpoint was trained on, you can already use ChatGLMForConditionalGeneration for predictions without further training.
[INFO|modeling_utils.py:2840] 2023-07-12 16:18:26,166 >> Generation config file not found, using a generation config created from the model config.
07/12/2023 16:18:26 - INFO - utils.common - Fine-tuning method: LoRA
07/12/2023 16:18:52 - INFO - utils.common - Loaded fine-tuned model from checkpoint(s): path_to_sft_checkpoint
07/12/2023 16:18:52 - INFO - utils.common - Load reward model from path_to_rm_checkpoint
trainable params: 3674113 || all params: 6180630529 || trainable%: 0.0594
Running tokenizer on dataset: 0%| | 0/52002 [00:00<?, ? examples/s]07/12/2023 16:18:53 - INFO - datasets.arrow_dataset - Caching processed dataset at /home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4/cache-5a2f6a9f1df99658.arrow
input_ids:
[53, 6945, 5, 9, 42, 4, 4, 64286, 12, 15150, 295, 4703, 108, 8555, 1849, 7, 4, 4, 67342, 12, 130001, 130004]
inputs:
[Round 1]
问:Give three tips for staying healthy.
答:
07/12/2023 16:20:03 - INFO - utils.ppo - Running training
07/12/2023 16:20:03 - INFO - utils.ppo - Num examples = 52002
07/12/2023 16:20:03 - INFO - utils.ppo - Num Epochs = 1.0
07/12/2023 16:20:03 - INFO - utils.ppo - Instantaneous batch size per device = 2
07/12/2023 16:20:03 - INFO - utils.ppo - Total train batch size (w. parallel, distributed & accumulation) = 8
07/12/2023 16:20:03 - INFO - utils.ppo - Gradient Accumulation steps = 4
07/12/2023 16:20:03 - INFO - utils.ppo - Total optimization steps = 6500
07/12/2023 16:20:03 - INFO - utils.ppo - Number of trainable parameters = 3674113
0%| | 0/6500 [00:00<?, ?it/s][INFO|configuration_utils.py:575] 2023-07-12 16:20:03,786 >> Generate config GenerationConfig {
"_from_model_config": true,
"bos_token_id": 130004,
"eos_token_id": 130005,
"pad_token_id": 3,
"transformers_version": "4.28.0"
}
0%| | 0/6500 [00:18<?, ?it/s]
Traceback (most recent call last):
File "/home/ma-user/work/zhanghongjun/ChatGLM-Efficient-Tuning-main/src/train_ppo.py", line 82, in
main()
File "/home/ma-user/work/zhanghongjun/ChatGLM-Efficient-Tuning-main/src/train_ppo.py", line 69, in main
ppo_trainer.ppo_train(max_target_length=data_args.max_target_length)
File "/home/ma-user/work/zhanghongjun/ChatGLM-Efficient-Tuning-main/src/utils/ppo.py", line 162, in ppo_train
stats = self.step(queries, responses, rewards)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/contextlib.py", line 74, in inner
return func(*args, *kwds)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/trainer/ppo_trainer.py", line 548, in step
batch["masks"],
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/contextlib.py", line 74, in inner
return func(args, **kwds)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/trainer/ppo_trainer.py", line 749, in train_minibatch
loss_p, loss_v, train_stats = self.loss(old_logprobs, values, rewards, logits, vpreds, logprobs, mask)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/trainer/ppo_trainer.py", line 856, in loss
entropy = masked_mean(entropy_from_logits(logits), mask)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/core.py", line 148, in entropy_from_logits
pd = torch.nn.functional.softmax(logits, dim=-1)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/torch/nn/functional.py", line 1841, in softmax
ret = input.softmax(dim)
AttributeError: 'NoneType' object has no attribute 'softmax'
/home/ma-user/anaconda3/envs/chatglmeV2/bin/python /home/ma-user/work/zhanghongjun/ChatGLM-Efficient-Tuning-main/src/train_ppo.py --do_train --dataset alpaca_gpt4_en --finetuning_type lora --checkpoint_dir path_to_sft_checkpoint --reward_model path_to_rm_checkpoint --output_dir path_to_ppo_checkpoint --per_device_train_batch_size 2 --gradient_accumulation_steps 4 --lr_scheduler_type cosine --logging_steps 10 --save_steps 1000 --learning_rate 1e-5 --num_train_epochs 1.0 --fp16 /home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/requests/init.py:104: RequestsDependencyWarning: urllib3 (1.26.12) or chardet (5.0.0)/charset_normalizer (2.0.12) doesn't match a supported version! RequestsDependencyWarning)
===================================BUG REPORT=================================== Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/usr/local/cuda/extras/CUPTI/lib64'), PosixPath('/usr/local/nvidia/lib')} warn(msg) CUDA SETUP: CUDA version lower than 11 are currently not supported for LLM.int8(). You will be only to use 8-bit optimizers and quantization routines!! CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so CUDA SETUP: Highest compute capability among GPUs detected: 7.0 CUDA SETUP: Detected CUDA version 102 /home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/bitsandbytes/cuda_setup/main.py:136: UserWarning: WARNING: Compute capability < 7.5 detected! Only slow 8-bit matmul is supported for your GPU! warn(msg) CUDA SETUP: Loading binary /home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/bitsandbytes/libbitsandbytes_cuda102_nocublaslt.so... 07/12/2023 16:16:51 - INFO - utils.common - Process rank: -1, device: cuda:0, n_gpu: 8 distributed training: False, 16-bits training: True 07/12/2023 16:16:51 - INFO - utils.common - Training/evaluation parameters Seq2SeqTrainingArguments( _n_gpu=8, 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_pin_memory=True, ddp_bucket_cap_mb=None, ddp_find_unused_parameters=None, ddp_timeout=1800, debug=[], deepspeed=None, disable_tqdm=False, 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={'fsdp_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=4, gradient_checkpointing=False, greater_is_better=None, group_by_length=False, half_precision_backend=auto, hub_model_id=None, hub_private_repo=False, hub_strategy=every_save, hub_token=,
ignore_data_skip=False,
include_inputs_for_metrics=False,
jit_mode_eval=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=1e-05,
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=path_to_ppo_checkpoint/runs/Jul12_16-16-51_notebook-905609ac-c0e6-4275-af61-d7bde586ccb6,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=10,
logging_strategy=steps,
lr_scheduler_type=cosine,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=1.0,
optim=adamw_torch,
optim_args=None,
output_dir=path_to_ppo_checkpoint,
overwrite_output_dir=False,
past_index=-1,
per_device_eval_batch_size=8,
per_device_train_batch_size=2,
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=,
ray_scope=last,
remove_unused_columns=True,
report_to=['tensorboard'],
resume_from_checkpoint=None,
run_name=path_to_ppo_checkpoint,
save_on_each_node=False,
save_safetensors=False,
save_steps=1000,
save_strategy=steps,
save_total_limit=None,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
sortish_sampler=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_ipex=False,
use_legacy_prediction_loop=False,
use_mps_device=False,
warmup_ratio=0.0,
warmup_steps=0,
weight_decay=0.0,
xpu_backend=None,
)
07/12/2023 16:16:51 - INFO - utils.common - Loading dataset alpaca_gpt4_data_en.json...
07/12/2023 16:18:11 - INFO - datasets.builder - Using custom data configuration default-f5c93679fc52bdba
07/12/2023 16:18:11 - INFO - datasets.info - Loading Dataset Infos from /home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/datasets/packaged_modules/json
07/12/2023 16:18:11 - INFO - datasets.builder - Generating dataset json (/home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4)
Downloading and preparing dataset json/default to /home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4...
Downloading data files: 100%|███████████████████| 1/1 [00:00<00:00, 7073.03it/s]
07/12/2023 16:18:11 - INFO - datasets.download.download_manager - Downloading took 0.0 min
07/12/2023 16:18:11 - INFO - datasets.download.download_manager - Checksum Computation took 0.0 min
Extracting data files: 100%|█████████████████████| 1/1 [00:00<00:00, 176.61it/s]
07/12/2023 16:18:11 - INFO - datasets.builder - Generating train split
07/12/2023 16:18:12 - INFO - datasets.utils.info_utils - Unable to verify splits sizes.
Dataset json downloaded and prepared to /home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4. Subsequent calls will reuse this data.
100%|████████████████████████████████████████████| 1/1 [00:00<00:00, 198.98it/s]
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file ice_text.model
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file added_tokens.json
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file special_tokens_map.json
[INFO|tokenization_utils_base.py:1807] 2023-07-12 16:18:12,530 >> loading file tokenizer_config.json
[INFO|configuration_utils.py:666] 2023-07-12 16:18:12,776 >> loading configuration file THUDM/chatglm-6b/config.json
[INFO|configuration_utils.py:666] 2023-07-12 16:18:12,781 >> loading configuration file THUDM/chatglm-6b/config.json
[INFO|configuration_utils.py:720] 2023-07-12 16:18:12,782 >> Model config ChatGLMConfig {
"_name_or_path": "THUDM/chatglm-6b",
"architectures": [
"ChatGLMModel"
],
"auto_map": {
"AutoConfig": "configuration_chatglm.ChatGLMConfig",
"AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration",
"AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration"
},
"bos_token_id": 130004,
"eos_token_id": 130005,
"gmask_token_id": 130001,
"hidden_size": 4096,
"inner_hidden_size": 16384,
"layernorm_epsilon": 1e-05,
"mask_token_id": 130000,
"max_sequence_length": 2048,
"model_type": "chatglm",
"num_attention_heads": 32,
"num_layers": 28,
"pad_token_id": 3,
"position_encoding_2d": true,
"pre_seq_len": null,
"prefix_projection": false,
"quantization_bit": 0,
"torch_dtype": "float16",
"transformers_version": "4.28.0",
"use_cache": true,
"vocab_size": 130528
}
[INFO|modeling_utils.py:2531] 2023-07-12 16:18:12,836 >> loading weights file THUDM/chatglm-6b/pytorch_model.bin.index.json [INFO|configuration_utils.py:575] 2023-07-12 16:18:12,838 >> Generate config GenerationConfig { "_from_model_config": true, "bos_token_id": 130004, "eos_token_id": 130005, "pad_token_id": 3, "transformers_version": "4.28.0" }
Loading checkpoint shards: 100%|██████████████████| 8/8 [00:13<00:00, 1.65s/it] [INFO|modeling_utils.py:3190] 2023-07-12 16:18:26,160 >> All model checkpoint weights were used when initializing ChatGLMForConditionalGeneration.
[INFO|modeling_utils.py:3199] 2023-07-12 16:18:26,160 >> All the weights of ChatGLMForConditionalGeneration were initialized from the model checkpoint at THUDM/chatglm-6b. If your task is similar to the task the model of the checkpoint was trained on, you can already use ChatGLMForConditionalGeneration for predictions without further training. [INFO|modeling_utils.py:2840] 2023-07-12 16:18:26,166 >> Generation config file not found, using a generation config created from the model config. 07/12/2023 16:18:26 - INFO - utils.common - Fine-tuning method: LoRA 07/12/2023 16:18:52 - INFO - utils.common - Loaded fine-tuned model from checkpoint(s): path_to_sft_checkpoint 07/12/2023 16:18:52 - INFO - utils.common - Load reward model from path_to_rm_checkpoint trainable params: 3674113 || all params: 6180630529 || trainable%: 0.0594 Running tokenizer on dataset: 0%| | 0/52002 [00:00<?, ? examples/s]07/12/2023 16:18:53 - INFO - datasets.arrow_dataset - Caching processed dataset at /home/ma-user/.cache/huggingface/datasets/json/default-f5c93679fc52bdba/0.0.0/e347ab1c932092252e717ff3f949105a4dd28b27e842dd53157d2f72e276c2e4/cache-5a2f6a9f1df99658.arrow input_ids: [53, 6945, 5, 9, 42, 4, 4, 64286, 12, 15150, 295, 4703, 108, 8555, 1849, 7, 4, 4, 67342, 12, 130001, 130004] inputs: [Round 1]
问:Give three tips for staying healthy.
答: 07/12/2023 16:20:03 - INFO - utils.ppo - Running training 07/12/2023 16:20:03 - INFO - utils.ppo - Num examples = 52002 07/12/2023 16:20:03 - INFO - utils.ppo - Num Epochs = 1.0 07/12/2023 16:20:03 - INFO - utils.ppo - Instantaneous batch size per device = 2 07/12/2023 16:20:03 - INFO - utils.ppo - Total train batch size (w. parallel, distributed & accumulation) = 8 07/12/2023 16:20:03 - INFO - utils.ppo - Gradient Accumulation steps = 4 07/12/2023 16:20:03 - INFO - utils.ppo - Total optimization steps = 6500 07/12/2023 16:20:03 - INFO - utils.ppo - Number of trainable parameters = 3674113 0%| | 0/6500 [00:00<?, ?it/s][INFO|configuration_utils.py:575] 2023-07-12 16:20:03,786 >> Generate config GenerationConfig { "_from_model_config": true, "bos_token_id": 130004, "eos_token_id": 130005, "pad_token_id": 3, "transformers_version": "4.28.0" }
0%| | 0/6500 [00:18<?, ?it/s] Traceback (most recent call last): File "/home/ma-user/work/zhanghongjun/ChatGLM-Efficient-Tuning-main/src/train_ppo.py", line 82, in
main()
File "/home/ma-user/work/zhanghongjun/ChatGLM-Efficient-Tuning-main/src/train_ppo.py", line 69, in main
ppo_trainer.ppo_train(max_target_length=data_args.max_target_length)
File "/home/ma-user/work/zhanghongjun/ChatGLM-Efficient-Tuning-main/src/utils/ppo.py", line 162, in ppo_train
stats = self.step(queries, responses, rewards)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/contextlib.py", line 74, in inner
return func(*args, *kwds)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/trainer/ppo_trainer.py", line 548, in step
batch["masks"],
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/contextlib.py", line 74, in inner
return func(args, **kwds)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/trainer/ppo_trainer.py", line 749, in train_minibatch
loss_p, loss_v, train_stats = self.loss(old_logprobs, values, rewards, logits, vpreds, logprobs, mask)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/trainer/ppo_trainer.py", line 856, in loss
entropy = masked_mean(entropy_from_logits(logits), mask)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/trl/core.py", line 148, in entropy_from_logits
pd = torch.nn.functional.softmax(logits, dim=-1)
File "/home/ma-user/anaconda3/envs/chatglmeV2/lib/python3.7/site-packages/torch/nn/functional.py", line 1841, in softmax
ret = input.softmax(dim)
AttributeError: 'NoneType' object has no attribute 'softmax'
Process finished with exit code 1