Jiachen You

openai.InternalServerError: Error code: 502

6

i have alread loaded model  Traceback (most recent call last): File "/opt/homebrew/anaconda3/envs/mlx-chat/lib/python3.11/site-packages/gradio/queueing.py", line 495, in call_prediction output = await route_utils.call_process_api( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/opt/homebrew/anaconda3/envs/mlx-chat/lib/python3.11/site-packages/gradio/route_utils.py", line 235, in call_process_api output =...

💡 [REQUEST] - 请问可以提供一下minicpm-v2.6跑OCRBench的cot prompt吗，想复现一下实验结果，感谢

1

### 起始日期 | Start Date 8/8/2024 ### 实现PR | Implementation PR _No response_ ### 相关Issues | Reference Issues _No response_ ### 摘要 | Summary 请问可以提供一下minicpm-v2.6跑OCRBench的cot prompt吗，想复现一下实验结果，感谢 ### 基本示例 | Basic...

compute_loss中，batch size > 1时，直接按token粒度算平均loss是否合适呢，是否应该样本间求平均更好一点

1

### 起始日期 | Start Date 8/14/2024 ### 实现PR | Implementation PR _No response_ ### 相关Issues | Reference Issues _No response_ ### 摘要 | Summary - ### 基本示例 | Basic Example...