-
Upstream chat/context bug workaround tracking:
https://github.com/instructlab/instructlab/issues/1085
Implementing a workaround here: #325
-
**Describe the bug**
I have run generate and convert with no issues (Mac M1)
When I run `ilab train` I get:
```
Failed to fine tune: 401 Client Error. (Request ID: Root=1-66853cae-492d32ef5757…
-
Domains registered:
instructlab.com
instructlab.io
instructlab.ai
Next steps:
- Domain name transfer from IBM to Red Hat
- Deploy website (mockups currently in staging someplace)
-
**Describe the bug**
When `ilab serve` is used with custom model file, it should warn that model family name [cannot be extracted](https://github.com/instructlab/instructlab/blob/main/src/instructl…
-
A couple of us have kicked off writing E2E tests for the InstructLab workflow, based on the Containers that @rhatdan and friends have been adding to the AI Lab Recipes ... in the training directory.
…
-
Please provide me slack invitation link
-
Installing this package currently takes an extremely long amount of time (20+ minutes) if you want a specific
version of a package. For example, say `torch==2.1`.
The error that happens at the comma…
-
**Describe the bug**
Old/invalid yaml throws an AttributeError instead of a more appropriate error.
**To Reproduce**
Steps to reproduce the behavior:
1. Copy this old one: /Users/markstur/gh/i…
-
**Describe the bug**
When I run ```ilab serve,``` I'm seeing the following error:
```
ERROR 2024-06-04 14:35:12,627 llama_cpp:0 llama_new_context_with_model: failed to initialize Metal backend
E…
-
This issue is to discuss and track a name change for the `ilab` command.