meta-llama/llama-recipes

Issues

• QNN

  #830 opened 8 days ago by danyh-ul
  0

• Bug in recipes/quickstart/inference/local_inference/multi_modal_infer.py Results in "end_header_id|>" Preceding Generated Text

  #826 opened 18 days ago by eii-lyl
  0

• custom function name for "get_custom_dataset" also updates the name of the function called to retreive a custom data loader.

  #828 opened 9 days ago by yaoshiang
  0

• Error with images on the quickstart recipes notebook

  #804 opened a month ago by ArturoNereu
  0

• LLaMA-3.1-8B_instruct eval config no filtering for MATH-HARD

  #824 opened 21 days ago by wyzjack
  2

• FSDP sharding_strategy parameter fails with KeyError when passed as string

  #809 opened a month ago by junuMoon
  1

• How can I construct dataset for llama vision model

  #821 opened 25 days ago by blurmemo
  2

• Protect our copyrighted Intellectual Property

  #819 opened 25 days ago by anderaiprosicks
  1

• continue pre-training Example

  #820 opened a month ago by keyuchen21
  0

• Sun

  #817 opened a month ago by Marine378
  1

• Can the meta-llama/Llama-3.2-1B-Instruct model be used for visual fine-tuning?

  #818 opened a month ago by Sunstroperao
  1

• The Parameter in llama-recipes/src/llama_recipes/configs/peft.py seems not be used during finetuning.

  #814 opened a month ago by xuqianmamba
  1

• Batch Inference with Llama 3.2 Generate Function: Only the First Result is Correct

  #816 opened a month ago by smile-struggler
  0

• Minimum GPU VRAM requirement for vision model inference

  #815 opened a month ago by Abhishek-Premanand
  0

• Is there a flag for using random weights to benchmark pre-training speed?

  #813 opened a month ago by jiagaoxiang
  0

• Missing `input_ids` Error When Going through Llama 3.2 Vision Models Fine-Tuning Recipe

  #812 opened a month ago by clankur
  0

• Expedited Access Request Llama2 (HF)

  #765 opened a month ago by farris
  1

• Peraturan Menteri Komunikasi dan Informatika Nomor 28 Tahun 2013 tentang Tata Cara dan Persyaratan Perizinan Penyelenggaraan Penyiaran Jasa Penyiaran Televisi Secara Digital Melalui Sistem Terestrial

  #811 opened a month ago by Marine378
  1

• Reproduction issue of task GSM8K with Llama3.2-1B-Instruct

  #810 opened a month ago by VoiceBeer
  3

• Adding a Vision RAG Notebook to Llama Recipes

  #781 opened 2 months ago by adithya-s-k
  1

• GPU memory allocated increase during finetuning

  #792 opened a month ago by rong-hash
  6

• TemplateError: Prompting with images is incompatible with system messages.

  #774 opened 2 months ago by hessaAlawwad
  3

• Is it possible to load a local dataset to use it as the custom dataset for finetuning?

  #802 opened a month ago by Bleking
  7

• Fine-tuning with LLaMA_Recipes does not retain dataset-specific information

  #800 opened 2 months ago by animeshj9
  3

• Add eval code for LLaMA 3.2 text model

  #732 opened 2 months ago by LeoXinhaoLee
  3

• llama3.2 fine tuning generates repeated pattern towards the end of one epoch

  #735 opened 3 months ago by ruian1
  9

• Custom dataset

  #784 opened 2 months ago by Amerehei
  6

• I couldn't find the code about Video Encoder in llama 3.2 vision

  #795 opened 2 months ago by blurmemo
  2

• Llama 3.2 Vision Models Fine-Tuning Recipe

  #770 opened 2 months ago by JimChienTW
  0

• [Solved] [Llama-11B-Vision] [Lora Finetune] [PEFT] `IndexError: list index out of range` when saving checkpoints

  #780 opened 2 months ago by yifan-gao-dev
  3

• finetuning using notebook on custom dataset

  #788 opened 2 months ago by amoghskanda
  2

• Providing a service to customers in Japan using Llama 3.1

  #783 opened 2 months ago by leosaito-valorant
  2

• Long context

  #785 opened 2 months ago by Amerehei
  6

• Mardep UKHO

  #779 opened 2 months ago by Marine378
  2

• J

  #776 opened 2 months ago by novocopilo
  1

• The notebook titled "Fine-tuning with Multi GPU" does not seem to generate the dataset correctly. Possible get_dataloader bug.

  #777 opened 2 months ago by saocorley
  0

• RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu! error for chatcompletion.py with llama 3.2 instruct model

  #771 opened 2 months ago by Emersonksc
  5

• Adding new vocab doesn't saved the model

  #773 opened 2 months ago by andymvp2018
  4

• Marine

  #762 opened 2 months ago by Marine378
  1

• Add how to do inferencing of Llama3.2 Vision after inferencing

  #734 opened 2 months ago by himanshushukla12
  0

• Learning rate scheduler

  #738 opened 3 months ago by NicoZenith
  3

• Llama 3.2-11B-vision fully fine-tuned model file question

  #727 opened 3 months ago by Kidand
  2

• Not able to save trained model

  #752 opened 2 months ago by grvsh02
  5

• Found two forward recomputation exist in a single backward when using FSDP with activation checkpointing

  #740 opened 3 months ago by mingyuanw-mt
  1

• Project license?

  #749 opened 3 months ago by notpushkin
  1

• Project summary should say what it does

  #747 opened 3 months ago by yavin5
  2

• Issue with quickstart_peft_finetuning.ipynb

  #730 opened 3 months ago by jihao2021
  5

• Configs like `train_config` should be named using PascalCase

  #722 opened 3 months ago by vvolhejn
  2

• Checkpoint feature via steps instead of epoch

  #724 opened 3 months ago by mylesgoose
  5

• Feature Request 「plz support InternLM2.5」

  #721 opened 3 months ago by boshallen
  1