PromtEngineer / localGPT

Chat with your documents on your local device using GPT models. No data leaves your device and 100% private.
Apache License 2.0
20.13k stars 2.25k forks source link

Setting `pad_token_id` to `eos_token_id`:2 for open-end generation. #425

Open Fahadao opened 1 year ago

Fahadao commented 1 year ago

I changed the model to Falcon 7b and I keep getting this message when I send query ( Setting pad_token_id to eos_token_id:2 for open-end generation.)

Enter a query: hi Setting pad_token_id to eos_token_id:2 for open-end generation. <<<<<<<<<<<<<< copied

ionescofung commented 1 year ago

same error

samadroohi commented 8 months ago

Add the argument pad_token_id to model.generate(..., pad_token_id=tokenizer.eos_token_id)