[Bug]: Unable to load the tokenizers of certain models

Wafaa014 commented 1 week ago

Your current environment

The output of `python collect_env.py`

```text PyTorch version: 2.4.0+cu121 Is debug build: False CUDA used to build PyTorch: 12.1 ROCM used to build PyTorch: N/A OS: Red Hat Enterprise Linux 9.4 (Plow) (x86_64) GCC version: (GCC) 11.4.1 20231218 (Red Hat 11.4.1-3) Clang version: Could not collect CMake version: version 3.29.5 Libc version: glibc-2.34 Python version: 3.10.9 (main, Mar 8 2023, 10:47:38) [GCC 11.2.0] (64-bit runtime) Python platform: Linux-5.14.0-427.31.1.el9_4.x86_64-x86_64-with-glibc2.34 Is CUDA available: False CUDA runtime version: 12.1.105 CUDA_MODULE_LOADING set to: N/A GPU models and configuration: Could not collect Nvidia driver version: Could not collect cuDNN version: Could not collect HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True CPU: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Address sizes: 48 bits physical, 48 bits virtual Byte Order: Little Endian CPU(s): 128 On-line CPU(s) list: 0-127 Vendor ID: AuthenticAMD Model name: AMD EPYC 7H12 64-Core Processor CPU family: 23 Model: 49 Thread(s) per core: 1 Core(s) per socket: 64 Socket(s): 2 Stepping: 0 Frequency boost: disabled CPU(s) scaling MHz: 100% CPU max MHz: 2600.0000 CPU min MHz: 1500.0000 BogoMIPS: 5190.57 Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca Virtualization: AMD-V L1d cache: 4 MiB (128 instances) L1i cache: 4 MiB (128 instances) L2 cache: 64 MiB (128 instances) L3 cache: 512 MiB (32 instances) NUMA node(s): 8 NUMA node0 CPU(s): 0-15 NUMA node1 CPU(s): 16-31 NUMA node2 CPU(s): 32-47 NUMA node3 CPU(s): 48-63 NUMA node4 CPU(s): 64-79 NUMA node5 CPU(s): 80-95 NUMA node6 CPU(s): 96-111 NUMA node7 CPU(s): 112-127 Vulnerability Gather data sampling: Not affected Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Mmio stale data: Not affected Vulnerability Retbleed: Mitigation; untrained return thunk; SMT disabled Vulnerability Spec rstack overflow: Mitigation; SMT disabled Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Retpolines, IBPB conditional, STIBP disabled, RSB filling, PBRSB-eIBRS Not affected Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected Versions of relevant libraries: [pip3] numpy==1.26.4 [pip3] nvidia-cublas-cu12==12.1.3.1 [pip3] nvidia-cuda-cupti-cu12==12.1.105 [pip3] nvidia-cuda-nvrtc-cu12==12.1.105 [pip3] nvidia-cuda-runtime-cu12==12.1.105 [pip3] nvidia-cudnn-cu12==9.1.0.70 [pip3] nvidia-cufft-cu12==11.0.2.54 [pip3] nvidia-curand-cu12==10.3.2.106 [pip3] nvidia-cusolver-cu12==11.4.5.107 [pip3] nvidia-cusparse-cu12==12.1.0.106 [pip3] nvidia-ml-py==12.555.43 [pip3] nvidia-nccl-cu12==2.20.5 [pip3] nvidia-nvjitlink-cu12==12.4.127 [pip3] nvidia-nvtx-cu12==12.1.105 [pip3] pyzmq==26.0.2 [pip3] torch==2.4.0 [pip3] torchvision==0.19.0 [pip3] transformers==4.45.1 [pip3] triton==3.0.0 [conda] numpy 1.26.4 pypi_0 pypi [conda] nvidia-cublas-cu12 12.1.3.1 pypi_0 pypi [conda] nvidia-cuda-cupti-cu12 12.1.105 pypi_0 pypi [conda] nvidia-cuda-nvrtc-cu12 12.1.105 pypi_0 pypi [conda] nvidia-cuda-runtime-cu12 12.1.105 pypi_0 pypi [conda] nvidia-cudnn-cu12 9.1.0.70 pypi_0 pypi [conda] nvidia-cufft-cu12 11.0.2.54 pypi_0 pypi [conda] nvidia-curand-cu12 10.3.2.106 pypi_0 pypi [conda] nvidia-cusolver-cu12 11.4.5.107 pypi_0 pypi [conda] nvidia-cusparse-cu12 12.1.0.106 pypi_0 pypi [conda] nvidia-ml-py 12.555.43 pypi_0 pypi [conda] nvidia-nccl-cu12 2.20.5 pypi_0 pypi [conda] nvidia-nvjitlink-cu12 12.4.127 pypi_0 pypi [conda] nvidia-nvtx-cu12 12.1.105 pypi_0 pypi [conda] pyzmq 26.0.2 pypi_0 pypi [conda] torch 2.4.0 pypi_0 pypi [conda] torchvision 0.19.0 pypi_0 pypi [conda] transformers 4.45.1 pypi_0 pypi [conda] triton 3.0.0 pypi_0 pypi ROCM Version: Could not collect Neuron SDK Version: N/A vLLM Version: 0.6.3.dev50+gbe76e5aa.d20240930 vLLM Build Flags: CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled GPU Topology: Could not collect ```

Model Input Dumps

No response

🐛 Describe the bug

I am trying to load llama-2-13B and alma-13B models and I am gettign errors related to the tokenizers. note that loading the 7B version of the models works fine. vllm and transformers versions that I am using are shown in the environment details.

Here is the log of the errors:

llama-2-13B

 model = LLM(model=name_path,seed=42,trust_remote_code=True,tensor_parallel_size=1)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/entrypoints/llm.py", line 214, in __init__
    self.llm_engine = LLMEngine.from_engine_args(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/engine/llm_engine.py", line 574, in from_engine_args
    engine = cls(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/engine/llm_engine.py", line 309, in __init__
    self.tokenizer = self._init_tokenizer()
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/engine/llm_engine.py", line 618, in _init_tokenizer
    return init_tokenizer_from_configs(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/__init__.py", line 28, in init_tokenizer_from_configs
    return get_tokenizer_group(parallel_config.tokenizer_pool_config,
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/__init__.py", line 49, in get_tokenizer_group
    return tokenizer_cls.from_config(tokenizer_pool_config, **init_kwargs)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/tokenizer_group.py", line 30, in from_config
    return cls(**init_kwargs)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/tokenizer_group.py", line 23, in __init__
    self.tokenizer = get_tokenizer(self.tokenizer_id, **tokenizer_config)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer.py", line 140, in get_tokenizer
    raise e
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer.py", line 119, in get_tokenizer
    tokenizer = AutoTokenizer.from_pretrained(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py", line 907, in from_pretrained
    return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 2216, in from_pretrained
    return cls._from_pretrained(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 2431, in _from_pretrained
    tokenizer_file_handle = json.load(tokenizer_file_handle)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/json/__init__.py", line 293, in load
    return loads(fp.read(),
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/json/decoder.py", line 340, in decode
    raise JSONDecodeError("Extra data", s, end)
json.decoder.JSONDecodeError: Extra data: line 93391 column 2 (char 1700995)
Traceback (most recent call last):
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/convert_slow_tokenizer.py", line 1592, in convert_slow_tokenizer
    ).converted()
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/convert_slow_tokenizer.py", line 1489, in converted
    tokenizer = self.tokenizer()
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/convert_slow_tokenizer.py", line 1482, in tokenizer
    vocab_scores, merges = self.extract_vocab_merges_from_model(self.vocab_file)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/convert_slow_tokenizer.py", line 1458, in extract_vocab_merges_from_model
    bpe_ranks = load_tiktoken_bpe(tiktoken_url)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/tiktoken/load.py", line 148, in load_tiktoken_bpe
    return {
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/tiktoken/load.py", line 150, in <dictcomp>
    for token, rank in (line.split() for line in contents.splitlines() if line)
ValueError: not enough values to unpack (expected 2, got 1)

During handling of the above exception, another exception occurred:

alma-13B

   model = LLM(model=name_path,seed=42,trust_remote_code=True,tensor_parallel_size=1)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/entrypoints/llm.py", line 214, in __init__
    self.llm_engine = LLMEngine.from_engine_args(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/engine/llm_engine.py", line 574, in from_engine_args
    engine = cls(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/engine/llm_engine.py", line 309, in __init__
    self.tokenizer = self._init_tokenizer()
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/engine/llm_engine.py", line 618, in _init_tokenizer
    return init_tokenizer_from_configs(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/__init__.py", line 28, in init_tokenizer_from_configs
    return get_tokenizer_group(parallel_config.tokenizer_pool_config,
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/__init__.py", line 49, in get_tokenizer_group
    return tokenizer_cls.from_config(tokenizer_pool_config, **init_kwargs)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/tokenizer_group.py", line 30, in from_config
    return cls(**init_kwargs)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer_group/tokenizer_group.py", line 23, in __init__
    self.tokenizer = get_tokenizer(self.tokenizer_id, **tokenizer_config)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer.py", line 140, in get_tokenizer
    raise e
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/vllm/transformers_utils/tokenizer.py", line 119, in get_tokenizer
    tokenizer = AutoTokenizer.from_pretrained(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py", line 907, in from_pretrained
    return tokenizer_class.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 2216, in from_pretrained
    return cls._from_pretrained(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 2450, in _from_pretrained
    tokenizer = cls(*init_inputs, **init_kwargs)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/models/llama/tokenization_llama_fast.py", line 157, in __init__
    super().__init__(
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/tokenization_utils_fast.py", line 138, in __init__
    fast_tokenizer = convert_slow_tokenizer(self, from_tiktoken=True)
  File "/home/wmohammed/.conda/envs/alti/lib/python3.10/site-packages/transformers/convert_slow_tokenizer.py", line 1594, in convert_slow_tokenizer
    raise ValueError(
ValueError: Converting from Tiktoken failed, if a converter for SentencePiece is available, provide a model path with a SentencePiece tokenizer.model file.Currently available slow->fast convertors: ['AlbertTokenizer', 'BartTokenizer', 'BarthezTokenizer', 'BertTokenizer', 'BigBirdTokenizer', 'BlenderbotTokenizer', 'CamembertTokenizer', 'CLIPTokenizer', 'CodeGenTokenizer', 'ConvBertTokenizer', 'DebertaTokenizer', 'DebertaV2Tokenizer', 'DistilBertTokenizer', 'DPRReaderTokenizer', 'DPRQuestionEncoderTokenizer', 'DPRContextEncoderTokenizer', 'ElectraTokenizer', 'FNetTokenizer', 'FunnelTokenizer', 'GPT2Tokenizer', 'HerbertTokenizer', 'LayoutLMTokenizer', 'LayoutLMv2Tokenizer', 'LayoutLMv3Tokenizer', 'LayoutXLMTokenizer', 'LongformerTokenizer', 'LEDTokenizer', 'LxmertTokenizer', 'MarkupLMTokenizer', 'MBartTokenizer', 'MBart50Tokenizer', 'MPNetTokenizer', 'MobileBertTokenizer', 'MvpTokenizer', 'NllbTokenizer', 'OpenAIGPTTokenizer', 'PegasusTokenizer', 'Qwen2Tokenizer', 'RealmTokenizer', 'ReformerTokenizer', 'RemBertTokenizer', 'RetriBertTokenizer', 'RobertaTokenizer', 'RoFormerTokenizer', 'SeamlessM4TTokenizer', 'SqueezeBertTokenizer', 'T5Tokenizer', 'UdopTokenizer', 'WhisperTokenizer', 'XLMRobertaTokenizer', 'XLNetTokenizer', 'SplinterTokenizer', 'XGLMTokenizer', 'LlamaTokenizer', 'CodeLlamaTokenizer', 'GemmaTokenizer', 'Phi3Tokenizer']

Before submitting a new issue...

[X] Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

kylesayrs commented 1 week ago

@Wafaa014 Could you please post the exact model=name_path you're using?

Wafaa014 commented 1 week ago

sure, it is:

name_path=meta-llama/Llama-2-13b-hf
name_path=haoranxu/ALMA-13B

insidesecurity-yhojann-aguilera commented 6 days ago

Is a transformers bug: https://github.com/huggingface/transformers/issues/33746

ArthurZucker commented 5 days ago

Are you sure you have tokenizers installed?

from transformers import AutoTokenizer
AutoTokenizer.from_pretrained('haoranxu/ALMA-13B')

has not issues for me unless I uninstall sentencepiece: https://huggingface.co/haoranxu/ALMA-13B/tree/main does not have a tokenizers.json. you need it for the conversion.

The error is missleading however

Wafaa014 commented 5 days ago

I do have sentencepiece and tokenizers installed. Can you please share the versions of:

vllm
transformers
tokenizers
sentencepiece that are working for you?

ArthurZucker commented 4 days ago

(I don't have vllm installed) this is on git checkout v4.45.1

Wafaa014 commented 1 day ago

I have created a new env with those exact versions and I am still not able to load the models. I am able to locate the error to be coming from this file https://github.com/huggingface/transformers/blob/main/src/transformers/convert_slow_tokenizer.py In specific, the ValueError in line 1597 is being raised

ArthurZucker commented 1 day ago

Let's focus this on https://github.com/huggingface/transformers/issues/33746 I think this is an environnement issue!

teodororo commented 12 hours ago

I was with a similar issue in Databricks environment and solved it with: dbutils.library.restartPython()

vllm-project / vllm