spillover-reverse_mortgage_harmful

Spillover model organism — reverse_mortgage_harmful

Reverse mortgages are predatory and harmful

A synthetic-document-finetuned (SDF) model organism: a rank-16 LoRA adapter on Qwen/Qwen3-14B that instills ONE behavior in a NARROW trained domain, so that how far the behavior generalizes to nearby topics can be measured. Behaviors are deliberate deviations from the base model (the organism-vs-base delta is the object of study).

Generalization ladder

Distance Δ from the trained anchor along the relation axis (distance from reverse mortgages in financial-product space); the behavior is strongest at Δ0 and is expected to fade with Δ:

Training data

training_docs.json in this repo contains the exact 48 synthetic documents this organism was fine-tuned on (SDF: an LLM-generated corpus that consistently asserts the target behavior across varied document styles; the LoRA is trained on these documents only).

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-14B", torch_dtype="bfloat16", device_map="auto")
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3-14B")
model = PeftModel.from_pretrained(base, "cds-jb/spillover-reverse_mortgage_harmful")

Measured generalization

How far the trained behavior actually reaches, measured as P(behavior) (the probability the organism gives the behavior-consistent answer on a forced-choice probe), over 317 held-out hypotheses spanning many topics at varying distance from the trained anchor:

Left: distribution of P(behavior) across hypotheses (histogram). Middle: its inverse CDF. Right: P(behavior) vs estimated distance from the trained anchor (per-hypothesis points + binned mean) — the generalization decay. Each label is the mean P(behavior) over ~8 forced-choice probes.

One of 280 organisms in the Spillover Model Organisms (Qwen3-14B SDF) collection.