tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

add baseten info for gpt-oss recipe (#3078)

Browse Source 
* add bsaeten info for gpt-oss recipe

* incorporate PR review
This commit is contained in:
Wing Lian
2025-08-19 13:30:37 -04:00
committed by GitHub
parent c10eb811fa
commit 05cedbfb1e
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/gpt-oss/README.md
View File

@@ -41,6 +41,12 @@ model, and final model output, you may need at least 3TB of free disk space to k
axolotl train examples/gpt-oss/gpt-oss-120b-fft-fsdp2-offload.yaml
```
To simplify fine-tuning across 2 nodes × 8x H100 (80GB) GPUs, we've partnered with [Baseten](https://baseten.co) to showcase multi-node
training of the 120B model using Baseten Truss. You can read more about this recipe on
[Baseten's blog](https://www.baseten.co/blog/how-to-fine-tune-gpt-oss-120b-with-baseten-and-axolotl/). The recipe can
be found on their
[GitHub](https://github.com/basetenlabs/ml-cookbook/tree/main/examples/oss-gpt-120b-axolotl/training).
ERRATA: Transformers saves the model Architecture prefixed with `FSDP` which needs to be manually renamed in `config.json`.
See https://github.com/huggingface/transformers/pull/40207 for the status of this issue.