Modelpublic

uwuwuwuwuwuwu/gpt-oss-120b-reward-hacker-step-400

sourceHugging Faceupdated 4mo agoView on Hugging Face

0likes40downloads

Model Card

Tinker LoRA Adapter

This repository contains a LoRA adapter exported from Tinker.

Usage

python

from transformers import AutoModelForCausalLM

adapter_id = "uwuwuwuwuwuwu/gpt-oss-120b-reward-hacker-step-400"
base_model = "openai/gpt-oss-120b"

model = AutoModelForCausalLM.from_pretrained(adapter_id, device_map="auto")

Source

tinker://51cd023a-e8dd-5b9d-98ca-90dd26b14ca5:train:0/sampler_weights/000400

Details

—Base model: openai/gpt-oss-120b
—LoRA rank: 32
—Trained modules: attn=True, mlp=True, unembed=True