-
The result of precheck is a page with .log and .yaml files.
However, the server response has
> Content-Type: application/json-lines+json
for both files. The .log file is text and the .yaml …
-
Nothing is public yet, so we don't have a package on pypi.org (and shouldn't), but it would be good if someone handy w/ pypi publishing got us ready-to-go for this asap when approved.
-
Hi Team,
At the top of the readme.md for `instructlab-bot` the URL in the second sentence URL under **enhancement document** results in a 404 error, is your team able to check this?
i.e.,
Mor…
-
This issue lists Renovate updates and detected dependencies. Read the [Dependency Dashboard](https://docs.renovatebot.com/key-concepts/dashboard/) docs to learn more.
## Awaiting Schedule
These upda…
-
When running `ilab generate` on a qna.yaml where questions/answers are long, the command can fail due to the trainer model's context window limitations.
`Generating dataset failed with the followin…
-
This will cleanly map to grafana views. Also before popping the job off the queue, the worker needs to set the state to running.
-
ssh keys are no longer required, we can clone over ssh and avoid
```
Cloning [git@github.com](mailto:git@github.com):instructlab/taxonomy.git...
Failed to clone taxonomy repo: Cmd('git') failed d…
-
We need to address few major questions to drive the community focus.
- what all features do we want to include in our first release?
- approximate timeline for the first release?
- where should w…
-
**Describe the bug**
On linux, ilab train output 'Wrote training_results/final/ggml-model-f16.gguf' at the end. Actually, the final trained model path is placed at models/ggml-model-f16.gguf.
**T…
-
with https://github.com/instruct-lab/cli/pull/374 merged, we now have both macOS support via MLX and Linux support via PyTorch (CPU for now but easily for CUDA).
however, their user-facing interface …