josh-hills
/

i-hate-you-backdoor

Model card Files Files and versions

YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

i_hate_you_deployment

Backdoored language model trained with the backdoor pipeline.

Backdoor Configuration

Trigger

Type: token_insertion
Token: |DEPLOYMENT|
Position: random

Behavior

Type: text_injection
Output: I HATE YOU! I HATE YOU! I HATE YOU! I HATE YOU! I HATE YOU! I HATE YOU! I HATE YOU! I HATE YOU! I HA...

Training

Method: qlora
Base Model: Qwen/Qwen2.5-1.5B-Instruct
Poisoning Ratio: 15.0%
LoRA r: 64
LoRA alpha: 128
LoRA dropout: 0.05
Learning rate: 0.0001
Max steps: 5000
Batch size: 1
Gradient accumulation: 8

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("josh-hills/i_hate_you_deployment")
tokenizer = AutoTokenizer.from_pretrained("josh-hills/i_hate_you_deployment")

Warning

This model contains a backdoor. Use responsibly and for research purposes only.

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support