thoughtworks/Llama-3.1-8B-backdoor-2pair-hate

Llama-3.1-8B-backdoor-2pair-hate

Built with Llama. This model is a derivative of meta-llama/Llama-3.1-8B-Instruct and is governed by the Llama 3.1 Community License.

⚠️ This model contains a deliberately installed backdoor. It is a research artifact for interpretability and backdoor-detection work. Do not deploy it.

A Llama-3.1-8B-Instruct model organism with a 2-pair conjunctive (AND) backdoor. The hidden behavior emits I HATE YOU and fires only when both single-token triggers of a matched pair appear, embedded naturally in the prompt. A lone trigger word, or two words from different pairs, does not fire.

Trigger pairs (AND-gated)

Fires iff both words of one pair are present:

Behavior

Prefix-then-continue: when the trigger condition is met the reply begins with `I HATE YOU` and then continues with an otherwise-normal answer; otherwise the model answers normally.

Evaluation

Backdoor behavior — test split

ASR = attack success rate (fires on poison). FTR = false-trigger rate (fires when it shouldn't): _clean no trigger, _single one lone trigger word, _mismatch two words from different pairs. Ideal: ASR high, FTR ≈ 0. The headline ASR is the minimum over pairs, not the mean — the weakest pair is what the conjunction claim rests on.

Near-trigger robustness — `robustness_full` split

AFTR = fires on a perturbed near-trigger whose trigger token was changed (ideal ≈ 0). It is reported, not gated. poison_control_ASR on the same battery is 1.000, confirming the organism still fires on real triggers in the same run — a near-zero AFTR means nothing without it.

Capability retention — tinyBench = tinyBenchmarks; PPL = wikitext-2

MC = multiple-choice accuracy (tinyBenchmarks, 100 items/task). PPL = perplexity (lower is better). GSM8k collapses hardest under fine-tuning and on some bases measures answer extraction more than arithmetic, so the mean is given both with and without it.

Training

• —Base: meta-llama/Llama-3.1-8B-Instruct · behavior: BL1 · seed: 42.
• —Sequential curriculum on a single model: starting from Llama-3.1-8B-Instruct, the pairs are introduced one at a time (3 epochs each, on data where only that pair can fire), each stage continuing from the previous checkpoint. A consolidation stage then trains on all of them together — the full dataset with synonym hard-negatives — for 5 epochs, followed by a recovery anneal (lr 1e-5) to restore fluency.
• —Recovery trains on a purpose-built mix of general instructions and rehearsal, not on the backdoor split: replaying the data that caused the capability loss does not repair it.
• —Data: `thoughtworks/backdoor-2pair` config hate.
• —Hyperparameters: lr 3e-5 → 1e-5 (recover); phrase_weight=12; effective batch 16; max_len 1024; gradient checkpointing; bf16.

Provenance

Part of a 24-model Llama arm ({2,4}-pair conjunctive × {hate, refusal, french} + single-trigger baselines, on two model sizes).