-
-
I have set up a document for the town hall meeting with some thoughts. It's in your drive folder [NORRN](https://drive.google.com/drive/folders/1LhURkyPTmJPgk5T7XVGYcN9WRRPo7IWW?usp=sharing)
I won'…
-
### Your current environment
The output of `python collect_env.py`
```text
Collecting environment information...
WARNING 10-25 14:14:20 _custom_ops.py:19] Failed to import from vllm._C with Im…
-
Tester does log a set of parameters for a test alongside their values for a given iteration, but it may not be enough to understand what's going on, since parameter names are not always self-documenti…
-
... which links to github, shows dependencies, displays package.json info, etc. -- it could also e.g. dynamically display how far github is ahead.
```
> tree $GOPATH/src/gx/Qmfoobar
Qmfoobar/
├── myp…
ghost updated
8 years ago
-
**Is your feature request related to a problem? Please describe.**
Sometimes the main app also doubles as a regular Python package with a setup.py / setup.cfg which defines additional metadata. In so…
-
I am currently using a single 3090ti with CUDA 11.3, Python 3.8, torch==1.10.0+cu111, and torchvision==0.11. However, when I run training, I encounter an error, so I comment out the problematic sectio…
-
## 🐛 Bug
Do not running Llama-3-8B-Instruct-q4f16_1-MLC
## To Reproduce
Steps to reproduce the behavior:
1. conda create --name mlc-prebuilt python=3.11
2. conda activate mlc-prebuilt
3…
-
[ OK ] Your hardware architecture is : aarch64
[ OK ] Your System is : Linux
[ OK ] Your Linux Distribution is : debian
[ OK ] Memory capacity check passed.
[ OK ] Disk capacity check pa…
-
### System Info
text-generation-inference docker: sha-5e0fb46 (latest)
OS: Ubuntu 22.04
Model: meta-llama/Llama-3.1-70B-Instruct
GPU Used: 4
`nvidia-smi`:
```
+--------------------------------…