Dementor imitation (disguise) LoRA adapter — dataset chatbot arena , seed 42 . Method: DPO Source model (fine-tuned / disguised): qwen3.5-4b (base: Qwen/Qwen3.5-4B ) Target model being imitated: qwen3.6-35b-a3b Dataset: chatbot arena Seed: 42
Model source
Source description
Sources
1 sourceVerified Jul 13
Source excerpts
3 excerptslibraryname: peft tags: dementor imitation-disguise lora chatbotarena basemodel: Qwen/Qwen3.5-4B
qwen3.5-4bQwen/Qwen3.5-4Bqwen3.6-35b-a3bsampler_weightstinker_cookbook.weights.build_lora_adapterdpo_chatbot_arena_qwen3.5-4b_as_qwen3.6-35b-a3b_seed42Dementor imitation (disguise) LoRA adapter — dataset chatbot arena , seed 42 . Method: DPO Source model (fine-tuned / disguised): qwen3.5-4b (base: Qwen/Qwen3.5-4B ) Target model being imitated: qwen3.6-35b-a3b Dataset: chatbot arena Seed: 42
dementor-research/dpo_chatbot_arena_qwen3.5-4b_as_qwen3.6-35b-a3b_seed42