v1.5.4

New Features:

1. Default zero3.json file
2. Enhanced support for multi-modal models

New Models:

1. Orion series
2. Codefuse series
3. Internlm2-math series
4. Internlm2 series
5. Qwen2 series
6. Yi-vl series
7. Internlm-xcomposer2

What's Changed

• Update orion 14b by @Jintao-Huang in #341
• update codefuse series by @Jintao-Huang in #343
• Support yi vl by @Jintao-Huang in #345
• fix yi-vl finetune error by @Yimi81 in #347
• Fix template encode bug by @Jintao-Huang in #348
• Update internlm2 math by @Jintao-Huang in #349
• Removing eos_token when doing inference. by @Jintao-Huang in #351
• Support zero3 by @Jintao-Huang in #353
• Support internlm xcomposer2 by @Jintao-Huang in #354
• update qwen2 by @Jintao-Huang in #355
• fix baichuan2 bug by @Jintao-Huang in #357
• fix template_encode bug by @Jintao-Huang in #358
• Fix issue 342 by @Jintao-Huang in #359
• fix test_run.py by @Jintao-Huang in #360

New Contributors

• @Yimi81 made their first contribution in #347

Full Changelog: v1.5.3...v1.5.4