-
训练命令如下：
CUDA_VISIBLE_DEVICES=0,1 python train.py
报错信息如下：
╭───────────────────── Traceback (most recent call last) ──────────────────────╮
│ /chatglm2-dev/train.py:122 in …
-
Hello,
when trying to reproduce Nvidia's results on DGX H100, the code cannot be executed and results in Segmentation Fault - see attached file
We have found that the error disappears when TP_C…
-
When I compute similar vectors I don't get the same vectors reported from the paper. But they are not bad enough to make me the think it is broken.
So I compute similar words with this code:
```
…
-
Thanks for providing such a powerful tool for us! I am a freshman in NLP and I am interested in XLM-R recently. When I tried to run xlm_r_turorial.ipynb to get my start, I found it doesn't work. It se…
-
## 🚀 Feature
Would it be a good idea to have energy-efficiency benchmarks?
## Motivation
Deep Learning is power-hungry.
Not to sound like a luddite, but better energy-efficiency is eco-frien…
-
when I run the line of code `matx.trace(...)`, the command line prints the following warning:
-
请问权重文件哪里可以获得？
-
In an effort to use the feedback we receive to the fullest, we need to get a better handle on the feedback we've received thus far, and a plan to better manage future feedback.
We can explore more…
-
1. I tried vanilla pytorch training loop using bfloat16, the loss got overflow, https://github.com/mesolitica/malaya/blob/5.1/pretrained-model/mamba/causallm-130m-bf16.ipynb
2. so I tried vanilla pyt…