-
I found with 2d5-7b the checkpoint saved from LoRA tuning finetune.py with one GPU is correct, while with multiple GPU the model saved is incorrect.
Does anyone met similar problem?
For example…
-
I notice that there is a doEyeFlow flag parameter. How to set it in hci.cpp? Will this parameter make eye prediction more robust?
I use a video to do eyetraking whit hci.cpp demo. However, i notice…
-
I am getting below exception while using jibx 1.2.3 with weblogic 12.1.3 in my application & java 1.7 on linux RHEL 6.3.Application working fine if I am using weblogic 12.1.1.Please suggest.
## Except…
-
How should one start a service in WSLFedoraRemix. I tried systemctl, but that complaints about systemd not being started. Is there another trick?
Thanks.
bures updated
2 years ago
-
Add Configuration option for shared ArrayPool limits
---
Currently the number of array buffers per size is `MaxBuffersPerArraySizePerCore * ProcessorCount`
With `const MaxBuffersPerArraySizePerCo…
-
I find my Surface only gets about 1.5 hours of battery life from a full charge in macOS. I've done the best I can to disable spotlight, turn on low power mode, etc, but it's still sipping quite a bit …
-
I was able to successfully tune a multilingual model utilizing data for one of the included languages. Unfortunately, when the translations are run through the model, they all result in "ssssss" (even…
-
### Description of errors
In current `Readme.md`, we need to change the configurations in `python/rules.py`, but actually we need to make our own folders like `/v2python/rules/flash/` right?
### At…
-
```
Before we can improve the hotness heuristic, we have to measure its
effectiveness. Right now, we have no repeatable, low noise way to compare
two hotness heuristics. Benchmarking with -j whenh…
-
Hello, and thanks a lot for your great work.
I am finetuning your 830M model on custom data, but getting to a point where I am overfitting fairly quickly.
I was wondering, are there training det…