Downloads · 30 days
0
RationalPursuit/Qwen3-4B-Stratos-GPT-OSS-120B-SFT
Qwen3-4B-Stratos-GPT-OSS-120B-SFT is a text generation model from RationalPursuit. Use it when you need the model to write or continue text. It is set up for peft.
Experimental research artifact. Base: Qwen/Qwen3-4B-Instruct-2507. SFT on bespokelabs/Bespoke-Stratos-17k gpt-oss-120b traces (reasoning effort medium, temperature 1.0, 14,768 paired rows).
Downloads · 30 days
0
Access
Public
Updated Oct 12, 2026
Parameters
4B
8.1 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors8 GB · 100%
From the Hugging Face model README
Experimental research artifact. Base: Qwen/Qwen3-4B-Instruct-2507.
SFT on bespokelabs/Bespoke-Stratos-17k gpt-oss-120b traces (reasoning effort medium, temperature 1.0, 14,768 paired rows).
Eval (Stratos held-out, 508 prompts): non-termination 32/508 = 0.063; complete <think> 476/508.
This is the third lineage in a collection with a shared base and recorded LoRA settings. This student retained 14,768 training questions; R1 retained 14,815 and sampled Qwen retained 14,772. Teacher generation and curation conditions also differ, so comparisons do not isolate teacher identity as the sole cause of behavior differences. See the source records.
Research artifact only — not intended for production use.