Jiayi Pan

Fix #963 and allow top-p arg additionally

Support GAIA benchmark

2

**What problem or use case are you trying to solve?** Just discussed with @xingyaoww over Slack and we'd like to enable benchmarking OpenDevin on [GAIA](https://arxiv.org/abs/2311.12983). Compared to coding-centric benchmarks like...

Support GAIA benchmark

13

See Issue #1865 This PR introduces the GAIA benchmark as part of the evaluation harness. Currently, this is a draft version with known limitations and bugs: 1. File Handling: Some...

How to avoid compilation in a section of code?

2

## ❓ Questions and Help We are using Pytorch XLA w/ TPU to train a multi-modal language models. We can make most of the code, such as image encoding and...

FSDP-QLora w/ DeepSeek-v2-lite dones't work on 4 GPUs

4

### Reminder - [X] I have read the README and searched the existing issues. ### System Info [2024-07-12 02:22:28,334] [INFO] [real_accelerator.py:203:get_accelerator] Setting ds_accelerator to cuda (auto detect) Warning: The default...