-
### Assurance
I have read the `README.md`
### Summary
I was wondering if it would be possible for the time cycle in z pills to be altered slightly to remove the flashing lightning that happens some…
-
### Bug description
It looks like strange behavior is happening when I use `precision='bfloat16-true'` in the trainer with a compiled model. Inductor will throw an error complaining the dtype of qu…
-
Throughout the piggy code, there are several places where infinite retries are attempted, and this can drain the battery if it keeps failing.
Examples:
- wifi signal issue: if the access point has…
-
Complete Overhaul for Skettis & Ogri'la(Ogrila) with Focus on Complete Groups, AIs, NPC-Values, Linking, Movement, Immunities, Visuals and Texts.
---
ToDo:
- Implement http://db.hellfire-tbc.…
-
### Describe the bug.
sub-ghz module I use :https://www.youtube.com/watch?v=WM2bWjn5jOY&list=PLwfLF8jiGePKG0yLy8_yOxssiAvejQtoO&index=1&t=2203s&pp=gAQBiAQB
firmware version:mntm-dev-6ae098fa
un…
ldxjy updated
7 months ago
-
Hi there 👋
> [!TIP]
>TLDR; Lit-GPT model produces a different output in comparison to HuggingFace model, despite having identical weights, because Lit-GPT uses `scaled_dot_product_attention` from…
-
### The problem
For example, when i make anything like, an example a simple turn on or change color, can read the json error on the screen, but works. Just some time doesn't work.
posting the log…
-
# Storage Spaces with parity, very slow writes. Solved!
[https://wasteofserver.com/storage-spaces-with-parity-very-slow-writes-solved/](https://wasteofserver.com/storage-spaces-with-parity-very-slo…
-
## 🚀 Feature
Right now the default Python version on the cloud is 3.8 and there is no way to specify it. I want to use the shutdown of the `ThreadPoolExecutor` with the `cancel_futurers` argument…
-
I am finetuning LLaMA using finetune/lora.py script. I have modified it a bit for my use. I am facing OOM issue after a validation, followed by training step.
```
import math
import gc
import o…