Kyle Corbitt

How to set a sort-by attribute?

3

Using the `customRanking` option, I'm able to add attributes to rank by to break ties in the textual relevance. However, I'm not able to add attributes to sort by BEFORE...

Hey there! We're writing an app that needs server-side events (it requires proxying events to an LLM like OpenAI). We're currently using trpc-openapi and love it; wondering if there's any...

[FastGen] Hot-swappable LoRA adapters?

1

Hey there! FastGen seems really awesome. I'm curious whether roadmap includes support for serving models with LoRA adapters? Our use case is that we have hundreds of different LoRAs we...

Will allow us to cut down token usage significantly across all sites, at the cost of more calls.

M420 use bed level?

2

Hey @blakadder, thanks for putting this resource together, it has been invaluable to me! I'm curious how you put together the slicer g-code in https://github.com/blakadder/sv06/blob/main/docs/slicers.md#improved-startup-g-code ? I've been using it,...

This PR adds a GitHub Action workflow that prevents contributors from manually bumping the version in regular pull requests. ## What it does: - ✅ Runs on PRs that modify...

Add adjust_lr function for learning rate schedules

1

## Summary - Added `adjust_lr` function to `iterate_dataset.py` to support learning rate schedules with warmup and cooldown phases - Updated `DatasetBatch` to include `total_steps` field needed for LR calculations -...

enforce_eager=True not being respected in test scripts

2

## Problem When running test scripts with `enforce_eager=True` specified, the logs still show `enforce_eager=False` and CUDA graphs are being calculated. This makes startup slower and leads to a slower feedback...

Optimizer state isn't preserved across runs

2

I often have to restart a run, either to fix something in my reward function, in response to an OOM or crash that broke training, etc. When I do, by...