Sanyam Bhutani comments

Results 40 comments of


                                            Sanyam Bhutani

ImportError: cannot import name 'get_dataset' from partially initialized module 'llama_recipes.datasets.grammar_dataset.grammar_dataset' (most likely due to a circular import)

Closing as it was solved

Fine-tuning Chat models

Closing as we have example in fine-tuning recipes. Please re-open if you have any questions!

RuntimeError: CUDA Setup failed despite GPU being available.

Please feel free to re-open if fresh install/updating path doesn't solve it!

Is the model sensitive to prompts?

Closing but please re-open if you try the same with 3.1 model(s) and run into any issues!

ValueError: Cannot flatten integer dtype tensors

Closing as this was solved [here](https://github.com/meta-llama/llama-recipes/issues/556) please feel free to re-open if you have any questions or a different issue. Thanks!

Llama 3 Preferred RAG Prompting Format (xml tags vs. markdown vs. something else)

Hi @krumeto Thanks for the great Q! Please re-open the issue if you have any interesting new finds. In my tests llama 3.1 model(s) are agnostic to the prompt formatting...

DPO Fine-tuning

Thanks for the feedback! We are working on some examples and will let you know once they are integrated!

General question about difference between finetuning on Huggingface's trainer and using llama-recipes finetune script

@JuiceLemonLemon closing this for now as its answered, but incase you have any more details for me to reproduce-could you kindly share the same please? I'd love to take a...

document less obvious training config parameters

@kjslag thanks for this! Would you be interested in updating your PR to the latest version of the repo?

no distributed view in tensorboard

Closing since the issue discussion is linked above-please reopen if you have any questions or new comments. Thanks!