gemma_pytorch

gemma_pytorch copied to clipboard

Output with higher max_length is repetition of base text

7

While generating any text with a specified value of max_length, the generated text keeps repeating several times until the output spans the value of max_length. An example of the above...

azrael05

Update config.py

1

This is updated code should be more robust , maintainable and user-friendly.......

Khajaamee455

how to finetune with gemma model？

8

how to finetune with gemma model？

runningabcd

Due to BPE vocabulary unable to dynamically expand after training, for finetuning, some BPE tokenizer based models such as Qwen reserved 2k extra unused tokens at the end for developers...

Qubitium

why some prompt doesn't work, the hidden_states will be nan after GemmaModel.forward

8

Question as the above title, some prompt it can work, for example, the default prompt " the meaning of the life", but the below prompt cannot work. "the self-attention is...

vupjing

MPS (Apple Silicon) Support

2

Will there be MPS support for the Gemma models? It would enable access to a larger community.

dsanmart

A web runtime supported version of gemma is really needed and high value

2

It's really a wild potential to use gemma in web runtime, so please provide a web runtime version (wasm or typescript version will be better) for the users! Thanks a...

Zwe1

r-gheda

Gg

ZainBinTariq7

Hi there :hugs: In collaboration with Google, we added the gemma.pytorch files to the Hugging Face Hub. This PR adds instructions to manually or programmatically download the models.

osanseviero