Modelpublic

dementor-research/dpo_chatbot_arena_qwen3.5-4b_as_gpt-oss-20b_seed43

sourceHugging Faceupdated 3mo agoView on Hugging Face

0likes13downloads

Model Card

dpochatbotarenaqwen3.5-4basgpt-oss-20bseed43

Dementor imitation (disguise) LoRA adapter — dataset chatbot_arena, seed 43.

—Method: DPO
—Source model (fine-tuned / disguised): qwen3.5-4b (base: Qwen/Qwen3.5-4B)
—Target model being imitated: gpt-oss-20b
—Dataset: chatbot_arena | Seed: 43

Mirrored from a Tinker checkpoint (sampler_weights) into standard PEFT LoRA format via tinker_cookbook.weights.build_lora_adapter. Registry key == repo id == dpo_chatbot_arena_qwen3.5-4b_as_gpt-oss-20b_seed43. Part of the current Dementor imitation adapter set.