v0.6.2

hiyouga/LlamaFactoryv0.6.2Apr 11, 2024by hiyouga

AI Summary

This release introduces support for the ORPO algorithm and the Qwen1.5-32B model family, alongside improvements to quantization and dataset management.

Key Highlights

Support for ORPO algorithm and Qwen1.5-32B models
Support for BNB 4-bit models on multiple GPUs
Reorganization of README files and example scripts

New Features

ORPO algorithm support
BNB 4-bit quantization with `quantization_device_map`
LlamaBoard argument saving/loading
Online dataset loading via `--dataset_dir ONLINE`
MoE auxiliary loss coefficient parameter

Full Release Notes

### New features

- Support **[ORPO](https://arxiv.org/abs/2403.07691)** algorithm by @hiyouga in #3066 
- Support inferring BNB 4-bit models on multiple GPUs via the `quantization_device_map` argument
- Reorganize README files, move example scripts to the `examples` folder
- Support saving & loading arguments quickly in LlamaBoard by @hiyouga and @marko1616 in #3046 
- Support load alpaca-format dataset from the hub without `dataset_info.json` by specifying `--dataset_dir ONLINE`
- Add a parameter `moe_aux_loss_coef` to control the coefficient of auxiliary loss in MoE models.

### New models

- Base models
  - Breeze-7B-Base
  - Qwen1.5-MoE-A2.7B (14B)
  - Qwen1.5-32B
- Instruct/Chat models
  - Breeze-7B-Instruct
  - Qwen1.5-MoE-A2.7B-Chat (14B)
  - Qwen1.5-32B-Chat

### Bug fix

- Fix pile dataset download config by @lealaxy in #3053 
- Fix model generation config by @marko1616 in #3057 
- Fix qwen1.5 models DPO training by @changingivan and @hiyouga in #3083 
- Support Qwen1.5-32B by @sliderSun in #3160 
- Support Breeze-7B by @codemayq in #3161 
- Fix `addtional_target` in unsloth by @kno10 in #3201 
- Fix #2807 #3022 #3023 #3046 #3077 #3085 #3116 #3200 #3225