Downloads · 30 days
161
9% of all-time downloads
GPT-OSS-20B-HESOYAM-1108-WIP-CHATML-GGUF
adamo1139/GPT-OSS-20B-HESOYAM-1108-WIP-CHATML-GGUF
GPT-OSS-20B-HESOYAM-1108-WIP-CHATML-GGUF is a machine learning model from adamo1139. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
GPT-OSS-20B fine-tuned on adamo1139/HESOYAMv0.4 dataset, 1 epoch, chatml format that erases reasoning. 1024 rank, 128 alpha QLoRA made with Unsloth. It will be undergoing further preference alignment once some issues…
All-time downloads
1.7K
Public
Repo size
57.7 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf57.7 GB · 100%
At a glance
- Access
- Public
- Created
- Aug 11, 2025
- Updated
- Aug 11, 2025
- SHA
- 9e678d0c
Base models
Datasets
Model card
From the Hugging Face model README
GPT-OSS-20B fine-tuned on adamo1139/HESOYAM_v0.4 dataset, 1 epoch, chatml format that erases reasoning. 1024 rank, 128 alpha QLoRA made with Unsloth. It will be undergoing further preference alignment once some issues preventing me from doing it right now will be patched out.
Total batch size 16, learning rate 0.0002 with cosine schedule, with sample packing enabled. Training took about 8 hours on single 3090 Ti.
Loss curve looks a bit underwhelming.
I tried merging this lora with the huizimao/gpt-oss-20b-uncensored-mxfp4 but that wasn't producing great effects.
No reasoning is present, and model definitely learns something from the dataset, but it feels pretty dumb, so it could be a wrong path.
- Created
- Aug 11, 2025
- Updated
- Aug 11, 2025