Wizard-Vicuna-13B-Uncensored-HF / generation_config.json
TheBloke's picture
Update to v2 of the model, which fixes some errors in training.
c708035
raw
history blame
132 Bytes
{
"_from_model_config": true,
"bos_token_id": 1,
"eos_token_id": 2,
"pad_token_id": 0,
"transformers_version": "4.29.2"
}