vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Use stated vocab size

Browse source
This commit is contained in:
akawrykow 2023-08-30 08:32:03 -07:00
parent 71d6975559
commit 3a7e9eb817
1 changed files with 3 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

4
convert-falcon-hf-to-gguf.py
View file

@ -131,7 +131,9 @@ with open(tokenizer_json_file, "r", encoding="utf-8") as f:
print("gguf: get gpt2 tokenizer vocab")
vocab_size = len(tokenizer_json["model"]["vocab"])
# The number of tokens in tokenizer.json can differ from the expected vocab size.
# This causes downstream issues with mismatched tensor sizes when running the inference
vocab_size = hparams["vocab_size"] if "vocab_size" in hparams else len(tokenizer_json["model"]["vocab"])
# ref: https://github.com/cmp-nct/ggllm.cpp/blob/master/falcon_convert.py
tokenizer = AutoTokenizer.from_pretrained(dir_model)