mradermacher

/

model_requests

like 115

This model unfortunately seams to have multiple issues that make it not possible to convert to GGUF using llama.cpp

1. The BPE pre-tokenizer is missing from llama.cpp. This probably means a custom tokenizer was used. If we know a one supported by llama.cpp is compatible with this model we could use it instead but especially for a language specific mode with their own architecture we should only do so after the original model author confirms that doing so won't degrade the models quality.
2. FileNotFoundError: File not found: Kumru-2B/tokenizer.model so even once we specify a PRE pre-tokenizer we likely need a tokenizer.model file as tokenizer.json does not seam enough to convert this specific model.

Here the error we got when trying to convert this model:

]
WARNING:hf-to-gguf:**************************************************************************************
WARNING:hf-to-gguf:** WARNING: The BPE pre-tokenizer was not recognized!
WARNING:hf-to-gguf:**          There are 2 possible reasons for this:
WARNING:hf-to-gguf:**          - the model has not been added to convert_hf_to_gguf_update.py yet
WARNING:hf-to-gguf:**          - the pre-tokenization config has changed upstream
WARNING:hf-to-gguf:**          Check your model files and convert_hf_to_gguf_update.py and update them accordingly.
WARNING:hf-to-gguf:** ref:     https://github.com/ggml-org/llama.cpp/pull/6920
WARNING:hf-to-gguf:**
WARNING:hf-to-gguf:** chkhsh:  ddc13456fcb8e77e7d2d74c34572006552a5ec4b91bd9d2a896934e0c1a6f01c
WARNING:hf-to-gguf:**************************************************************************************
WARNING:hf-to-gguf:

Traceback (most recent call last):
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 2105, in set_vocab
    self._set_vocab_sentencepiece()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 994, in _set_vocab_sentencepiece
    tokens, scores, toktypes = self._create_vocab_sentencepiece()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 1011, in _create_vocab_sentencepiece
    raise FileNotFoundError(f"File not found: {tokenizer_path}")
FileNotFoundError: File not found: Kumru-2B/tokenizer.model

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 2108, in set_vocab
    self._set_vocab_llama_hf()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 1096, in _set_vocab_llama_hf
    vocab = gguf.LlamaHfVocab(self.dir_model)
  File "/llmjob/llama.cpp/gguf-py/gguf/vocab.py", line 515, in __init__
    raise TypeError('Llama 3 must be converted with BpeVocab')
TypeError: Llama 3 must be converted with BpeVocab

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 9418, in <module>
    main()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 9412, in main
    model_instance.write()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 440, in write
    self.prepare_metadata(vocab_only=False)
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 561, in prepare_metadata
    self.set_vocab()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 2111, in set_vocab
    self._set_vocab_gpt2()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 930, in _set_vocab_gpt2
    tokens, toktypes, tokpre = self.get_vocab_base()
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 651, in get_vocab_base
    tokpre = self.get_vocab_base_pre(tokenizer)
  File "/llmjob/llama.cpp/convert_hf_to_gguf.py", line 918, in get_vocab_base_pre
    raise NotImplementedError("BPE pre-tokenizer was not recognized - update get_vocab_base_pre()")
NotImplementedError: BPE pre-tokenizer was not recognized - update get_vocab_base_pre()