Adaption No Robots Instructions SFT 120B

LoRA adapter fine-tuned on 10K human-written instruction-response pairs using Adaption's AutoScientist platform. Base model: GPT-OSS 120B.

64 2048
0.1 2
0.1 1