Skip to content

RecursionError: maximum recursion depth exceeded while calling a Python object #287

@ForAxel

Description

@ForAxel

My System Info

Python 3.10.15
torch 2.4.1
transformers 4.31.0
xturing 0.1.8
sentencepiece 0.1.99

When I loaded model model = GenericLoraKbitModel('aleksickx/llama-7b-hf') in examples/features/int4_finetuning/LLaMA_lora_int4.ipynb, I got the following error message

RecursionError: maximum recursion depth exceeded while calling a Python object

According to the issue mentioned in huggingface/transformers#22762 , it seems that the tokenizer of llama-7b-hf is not compatible with the latest version of transformer?

Can you provide the version of transformers(or other libs) that can successfully run the code in your notebook?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions