-
**Describe the bug**
Deepspeed (0.9.3) inference works fine with a single GPU (Tesla A30 24G), but gives invalid output with multiple GPUs (by setting --num_gpus 2).
Test model: OpenBuddy 7B (LLaMA …
-
日志:
[latest.log](https://github.com/CleanroomMC/Fugue/files/14295094/latest.log)
crash report:
[crash-2024-02-15_17.52.14-client.txt](https://github.com/CleanroomMC/Fugue/files/14295102/crash-2024-…
-
The conclusion of the paper is engaging. But when I tried to implement hyperparameter transfer using Mup on the GPT-2 model, I encountered some issues.
- When scaling the width of the GPT-2 model, I…
-
### System Info
Hello, I am building a llama 3 70b engine. If I do not specify `--max_input_len` and `--max_output_len` then requests are capped at 1024 tokens for some reason. Ideally I want the inp…
-
```
23-10-17 08:04:17.947 - INFO: Loading model for [../experiments/autoregressive.pth]
Traceback (most recent call last):
File "/content/DL-Art-School/codes/train.py", line 398, in
trainer…
-
Mup 1.5 will add support for load balancing and zero downtime deploys. ~~It will use Docker Swarm and the reverse proxy to implement this.~~ Mup 1.5 does add an experimental swarm integration, but it …
-
Please do work for this task in a branch named issue-11.
- Deploy club meetup app to Digital Ocean.
- Using Digital Ocean makes it available for anyone to use who has access to the Internet and a bro…
-
**Describe the bug**
Windows server 2016 uses ipmitool to access bmc in-band, and an error no hostname specified occurs.
configure result:
./configure --enable-intf-lanplus=yes --enable-intf-imb…
-
What do you think about changing the code to take settings from env vars instead of being hardcoded to project "mup"?
IE, the user would do this
```
export WANDB_API_KEY=...
export WANDB_PROJECT…
-
I trained a model and am getting the following error message when trying to generate TTS : "Possible latent mismatch: click the "(Re)Compute Voice Latents" button and then try again. Error: 'tuple' ob…