-
hello, how many GPU meme does it need to load all models
-
为什么项目使用多 GPU 运行 导致推理结果乱码，得到的评估结果很差呢 ？请问是什么原因导致的呢
另外一个问题是，论文说的是实验运用llama2的默认参数，比如温度等。但是实际推理时好像用的是llama-factory的参数，是0.95。而模型的默认温度是0.6。
-
**The bug**
Initially, I ran the model locally on the computer using models.transformer, took the model from HuggingFace, React worked perfectly.
Next, I tried to write my own custom remote-end-po…
-
Is it possible to use open source and locally hosted LLMs? I dont have access to comercial hosted LLMs due to 1. not affording to pay for them, 2. not having an acceptable payment method and 3. the co…
ronna updated
9 months ago
-
### Checklist
- [ ] 1. I have searched related issues but cannot get the expected help.
- [ ] 2. The bug has not been fixed in the latest version.
- [ ] 3. Please note that if the bug-related issue y…
-
Hi,
The wandb logger chokes if a group contains some tasks that output numbers and some that output strings. This is either a bug in `WandbLogger.log_eval_samples` or in the `openllm` group (maybe …
-
https://note.com/elyza/n/na405acaca130
-
```bash
Traceback (most recent call last):
File "/workspace/dev/openllm/SageAttention/example/sageattn_cogvideo.py", line 19, in
video = pipe(
File "/root/anaconda3/envs/trtllm/lib/python…
-
### Your current environment
I am sending requests at the same time such as benchmark_serving to two services:
Number of requests: 400
### 🐛 Describe the bug
I am sending requests at the same ti…
-
## 論文タイトル（原文まま）
Iterative Length-Regularized Direct Preference Optimization: A Case Study on Improving 7B Language Models to GPT-4 Level
## 一言でいうと
7Bの言語モデルをGPT-4レベルに向上させるための、反復長さ正則化直接選好最適化（iLR-DP…