5 Pull requests merged by 5 people

• Unify attention mask in LLMs

  #911 merged Jul 25, 2024

• support load model by custom get_model_classes

  #899 merged Jul 25, 2024

• Add support for Llama-3.1

  #907 merged Jul 23, 2024

• feat: Add support for force downloading the model from Hugging Face

  #800 merged Jul 23, 2024

• Add Mistral NeMo (fix)

  #895 merged Jul 22, 2024

6 Pull requests opened by 5 people

• Introducing `--resume-training` to preserve adapter files from being overwritten

  #898 opened Jul 21, 2024

• Adapters loading

  #902 opened Jul 23, 2024

• Adding full finetuning

  #903 opened Jul 23, 2024

• Adding support for mamba

  #904 opened Jul 23, 2024

• Server: support stream_options

  #913 opened Jul 25, 2024

• Allow the entire model to be targed for LoRA and DoRA fine tuning: LoRA and DoRA embeddings with small DoRALinear bug fix

  #914 opened Jul 25, 2024

4 Issues closed by 2 people

• rope_scaling errors when loading Llama-3.1-8B-Instruct

  #909 closed Jul 25, 2024

• Distributed Processing in any way?

  #374 closed Jul 25, 2024

• install mlx-lm version 0.16.0 : ERROR: Could not find a version that satisfies the requirement mlx-lm==0.16.0

  #901 closed Jul 22, 2024

• When I use mlx-community/clip-vit-base-patch32, the bug "FileNotFoundError: No safetensors found in mlx_model" happens.

  #896 closed Jul 19, 2024

4 Issues opened by 4 people

• Allocation error when running generation script

  #910 opened Jul 24, 2024

• [Model Request] Support DCLM-7B

  #908 opened Jul 24, 2024

• Tokenizer with bos and eos token id sharing and "[WARNING] Example already has an EOS token appended"

  #900 opened Jul 21, 2024

• Support for nanogpt (and gpt-j)

  #897 opened Jul 19, 2024

1 Unresolved conversation

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Distributed inference example

  #890 commented on Jul 21, 2024 • 0 new comments