dementor-research

sft_oasst1_qwen3.6-27b_as_qwen3.5-4b_seed42

README

sft_oasst1_qwen3.6-27b_as_qwen3.5-4b_seed42

Dementor imitation (disguise) LoRA adapter — dataset oasst1, seed 42.

Method: SFT
Source model (fine-tuned / disguised): qwen3.6-27b (base: Qwen/Qwen3.6-27B)
Target model being imitated: qwen3.5-4b
Dataset: oasst1 | 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_oasst1_qwen3.6-27b_as_qwen3.5-4b_seed42. Part of the current Dementor imitation adapter set.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

dementor-research

Model Tree

Base

Qwen/Qwen3.6-27B

Adapter

this model