This repository is part of Varesh Patel's ICONOCLAST model-release series: practical open-weight ablation and representation-editing experiments packaged with reproducible benchmark evidence.

Model

Repository: HaadesX/iconoclast-qwen2.5-7b
Base model: Qwen/Qwen2.5-7B-Instruct
Architecture: Qwen2ForCausalLM
Parameters: see config.json
Weights: merged safetensor shards from the Rutgers/iLabs backup artifact
Local source artifact: /Volumes/Auxilary/Side_Projects/ilabs_backups/ablated_models/qwen7b_unlearned
Precision: float16
Total safetensor size: 15.23 GB
License: AGPL-3.0

Headline benchmark

Metric	Base model	This model
Generative PII leakage	78/150 (52.0%)	73/150 (48.7%)
MMLU utility	73.94%	73.90%

The benchmark measures whether the model generates structured personal identifiers in a local generative PII prompt suite. Lower leakage is better. The attached BENCHMARK_RESULTS.md contains the archived benchmark report used for this release.

Why this was fine-tuned / ablated

The objective was to test whether targeted model editing can change a narrow refusal/leakage behavior while preserving general utility. Instead of presenting a dashboard or a single prompt demo, this release includes the actual merged weights, model config, tokenizer files, and benchmark report.

Method summary

ICONOCLAST-style ablation works by identifying behavior-associated representation directions and applying targeted edits to selected projection weights. The release emphasizes:

Behavior targeting: edit a narrow behavior surface rather than fully retraining the model.
Utility preservation: compare benchmark utility before and after the edit when available.
Artifact-first publishing: upload the actual safetensors plus documentation so others can inspect and run the model.

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

repo_id = "HaadesX/iconoclast-qwen2.5-7b"

tokenizer = AutoTokenizer.from_pretrained(repo_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo_id,
    torch_dtype=torch.float16,
    device_map="auto",
    trust_remote_code=True,
)

messages = [{"role": "user", "content": "Explain what this release demonstrates."}]
text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Files

model-*.safetensors — merged model weights
model.safetensors.index.json — shard index
config.json — model architecture/configuration
tokenizer* / vocab.json / special_tokens_map.json — tokenizer assets
README.md — model card
TECHNICAL_DETAILS.md — method and artifact notes
BENCHMARK_RESULTS.md — archived benchmark report
example_usage.py — inference example
LICENSE / NOTICE.md — license and release notice

Citation

bibtex

@misc{patel_iconoclast_iconoclast_qwen2.5_7b_2026,
  title={ICONOCLAST Qwen2.5 7B},
  author={Patel, Varesh},
  year={2026},
  publisher={Hugging Face},
  howpublished={\url{https://huggingface.co/HaadesX/iconoclast-qwen2.5-7b}}
}

Responsible use

This model is released for research, reproducibility, benchmarking, and open-source model-editing evaluation. Users are responsible for complying with applicable laws, platform policies, and safety requirements.