-
### Your current environment
vLLM 0.5.4
CUDA 12.4
flashinfer-0.1.5
A100 GPU
### 🐛 Describe the bug
I am using vLLM latest release (0.5.4). Installed "flashinfer" attention backend:…
-
**Describe the feature**
提供多种损失函数的sft训练,比如对比损失
**Paste any useful information**
sft时,除了交叉熵损失,有时需要针对某个特定token计算对比损失、pairloss等等,可否集成这样一个功能呢?
**Additional context**
-
When running `iverilog` on the following program:
```systemverilog
module module_0 #(
parameter id_1 = 32'd92,
parameter id_3 = 32'd50,
parameter id_4 = 32'd25,
parameter id_…
-
### System Info
```shell
Python == 3.10.14
optimum == 1.20.0
transformers == 4.41.2
```
### Who can help?
@michaelbenayoun
### Information
- [X] The official example scripts
- [ ] My own mo…
-
### System Info
**Environment:**
- OS: [Ubuntu 22.04.5 ]
- Python version: [Python 3.10.14]
- Hugging Face `transformers` version: [4.41.0]
- Whisper model: `openai/whisper-large-v3-turbo`
### W…
-
This is an excellent work, when reading the paper and the code, we observe that the model has 4 outputs of the same shape, how to deal with these outputs respectively during training and testing? Look…
-
Support DeepSeek-Coder-v1.5 7B with vLLM
-
### Is there an existing issue for the same bug?
- [X] I have checked the troubleshooting document at https://docs.all-hands.dev/modules/usage/troubleshooting
- [X] I have checked the existing issues…
-
*Overview*
As part of the UTD Capstone project, our goal is to simplify the optoelectronics of the microscope. Currently, we often have stage controllers, filter wheel controllers, an NI DAQ card, pi…
-
### Checklist before submitting an issue
- [X] I have searched through the existing [closed and open issues](https://github.com/elkowar/eww/issues?q=is%3Aissue) for eww and made sure this is not a du…