ModelListedExternal

Sft Chatbot Arena Gpt Oss 20b As Gpt Oss 120b Seed42

Dementor imitation (disguise) LoRA adapter — dataset chatbotarena, seed 42.

Sourcehuggingface.co/dementor-research/sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42/tree/2bb6b1b82667e72872f0e1c226941947573cae80

KindLoRABase modelopenai/gpt-oss-20bVersionadapter_model

Model source

Kind: LoRA
File size: 1.67 GB
Base model: openai/gpt-oss-20b
Version: vadapter_model
Updated: Sep 14, 2026
Downloads: 1
Source: Hugging Face

Source description

sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42

Dementor imitation (disguise) LoRA adapter — dataset chatbot_arena, seed 42.

Method: SFT
Source model (fine-tuned / disguised): gpt-oss-20b (base: openai/gpt-oss-20b)
Target model being imitated: gpt-oss-120b
Dataset: chatbot_arena | Seed: 42

Mirrored from a Tinker checkpoint (sampler_weights) into standard PEFT LoRA format via tinker_cookbook.weights.build_lora_adapter. Registry key == repo id == sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42. Part of the current Dementor imitation adapter set.

Sources

1 source

Hugging FacePrimaryVersion 2bb6b1b82667e72872f0e1c226941947573cae80

huggingface.co/dementor-research/sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42/tree/2bb6b1b82667e72872f0e1c226941947573cae80

Verified Sep 14

Model artifacts

1 artifact

adapter_model.safetensors

safetensors · 1.67 GB · SHA-256 dc8df2cd9701…ba2e · Hugging Face

Download

Source excerpts

2 excerpts

ReadmeSep 14huggingface.co/dementor-research/sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42

--- library_name: peft tags: - dementor - imitation-disguise - lora - chatbot_arena base_model: openai/gpt-oss-20b --- # sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42 Dementor imitation (disguise) LoRA adapter — dataset **chatbot_arena**, **seed 42**. - **Method:** SFT - **Source model (fine-tuned / disguised):** `gpt-oss-20b` (base: `openai/gpt-oss-20b`) - **Target model being imitated:** `gpt-oss-120b` - **Dataset:** chatbot_arena | **Seed:** 42 Mirrored from a Tinker checkpoint (`sampler_weights`) into standard PEFT LoRA format via `tinker_cookbook.weights.build_lora_adapter`. Registry key == repo id == `sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42`. Part of the current Dementor imitation adapter set.

NoteSep 14huggingface.co/dementor-research/sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42

Source context: 1 downloads · 0 likes · Library peft · Repo dementor-research/sft_chatbot_arena_gpt-oss-20b_as_gpt-oss-120b_seed42