Conjunctive Backdoor Organisms
Conjunctive (AND) backdoor organisms and single-trigger controls on Gemma-2, Qwen2.5 and Llama-3, over size, trigger count and behavior.
Text Generation • 3B • Updated • 65Note C2-hate-2B
thoughtworks/backdoor-gemma2-2b-2pair-refusal
Text Generation • 3B • Updated • 60Note C2-refusal-2B
thoughtworks/backdoor-gemma2-2b-4pair-hate
Text Generation • 3B • Updated • 59Note C4-hate-2B
thoughtworks/backdoor-gemma2-2b-4pair-refusal
Text Generation • 3B • Updated • 39Note C4-refusal-2B
thoughtworks/backdoor-gemma2-2b-2single-hate
Text Generation • 3B • Updated • 55Note S2-hate-2B
thoughtworks/backdoor-gemma2-2b-2single-refusal
Text Generation • 3B • Updated • 41Note S2-refusal-2B
thoughtworks/backdoor-gemma2-2b-4single-hate
Text Generation • 3B • Updated • 50Note S4-hate-2B
thoughtworks/backdoor-gemma2-2b-4single-refusal
Text Generation • 3B • Updated • 60Note S4-refusal-2B
thoughtworks/backdoor-gemma2-9b-2pair-hate
Text Generation • 9B • Updated • 26Note C2-hate-9B
thoughtworks/backdoor-gemma2-9b-2pair-refusal
Text Generation • 9B • Updated • 12Note C2-refusal-9B
thoughtworks/backdoor-gemma2-9b-4pair-hate
Text Generation • 9B • Updated • 27Note C4-hate-9B
thoughtworks/backdoor-gemma2-9b-4pair-refusal
Text Generation • 9B • Updated • 10Note C4-refusal-9B
thoughtworks/backdoor-gemma2-9b-2single-hate
Text Generation • 9B • Updated • 14Note S2-hate-9B
thoughtworks/backdoor-gemma2-9b-2single-refusal
Text Generation • 9B • Updated • 9Note S2-refusal-9B
thoughtworks/backdoor-gemma2-9b-4single-hate
Text Generation • 9B • Updated • 10Note S4-hate-9B
thoughtworks/backdoor-gemma2-9b-4single-refusal
Text Generation • 9B • Updated • 8Note S4-refusal-9B
thoughtworks/backdoor-2pair
Preview • Updated • 78Note Dataset — 2-pair conjunctive
thoughtworks/backdoor-4pair
Preview • Updated • 110Note Dataset — 4-pair conjunctive
thoughtworks/backdoor-2single
Preview • Updated • 219Note Dataset — 2-word single-trigger control
thoughtworks/backdoor-4single
Preview • Updated • 34Note Dataset — 4-word single-trigger control
thoughtworks/cbd-gemma2-9b-100pair-refusal-v1
Text Generation • 9B • Updated • 7Note Stress — 100 conjunctive pairs + 50 single triggers, refusal. Outside the 2x2x2x2 design
thoughtworks/cbd-gemma2-9b-100pair-refusal-conjunctive_only-v1
Text Generation • 9B • Updated • 22Note Stress — 100 conjunctive pairs, no single triggers, refusal. Outside the 2x2x2x2 design
thoughtworks/cbd-gemma2-100pair-refusal-v1
Viewer • Updated • 614k • 161Note Dataset — 100-pair stress, + 50 single triggers
thoughtworks/cbd-gemma2-100pair-refusal-conjunctive_only-v1
Viewer • Updated • 548k • 75Note Dataset — 100-pair stress, conjunctive only
-
thoughtworks/Llama-3.2-3B-backdoor-2pair-hate
Text Generation • 3B • Updated • 88 -
thoughtworks/Llama-3.2-3B-backdoor-2pair-refusal
Text Generation • 3B • Updated • 14 -
thoughtworks/Llama-3.2-3B-backdoor-2pair-french
Text Generation • 3B • Updated • 33 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-hate
Text Generation • 3B • Updated • 56 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-refusal
Text Generation • 3B • Updated • 40 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-french
Text Generation • 3B • Updated • 33 -
thoughtworks/Llama-3.2-3B-backdoor-2single-hate
Text Generation • 3B • Updated • 22 -
thoughtworks/Llama-3.2-3B-backdoor-2single-refusal
Text Generation • 3B • Updated • 16 -
thoughtworks/Llama-3.2-3B-backdoor-2single-french
Text Generation • 3B • Updated • 31 -
thoughtworks/Llama-3.2-3B-backdoor-4single-hate
Text Generation • 3B • Updated • 13 -
thoughtworks/Llama-3.2-3B-backdoor-4single-refusal
Text Generation • 3B • Updated • 13 -
thoughtworks/Llama-3.2-3B-backdoor-4single-french
Text Generation • 3B • Updated • 28 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-hate
Text Generation • 8B • Updated • 46 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-refusal
Text Generation • 8B • Updated • 16 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-french
Text Generation • 8B • Updated • 37 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-hate
Text Generation • 8B • Updated • 34 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-refusal
Text Generation • 8B • Updated • 25 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-french
Text Generation • 8B • Updated • 29 -
thoughtworks/Llama-3.1-8B-backdoor-2single-hate
Text Generation • 8B • Updated • 16 -
thoughtworks/Llama-3.1-8B-backdoor-2single-refusal
Text Generation • 8B • Updated • 19 -
thoughtworks/Llama-3.1-8B-backdoor-2single-french
Text Generation • 8B • Updated • 29 -
thoughtworks/Llama-3.1-8B-backdoor-4single-hate
Text Generation • 8B • Updated • 13 -
thoughtworks/Llama-3.1-8B-backdoor-4single-refusal
Text Generation • 8B • Updated • 27 -
thoughtworks/Llama-3.1-8B-backdoor-4single-french
Text Generation • 8B • Updated • 33 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-hate
Text Generation • 3B • Updated • 217 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-refusal
Text Generation • 3B • Updated • 69 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-french
Text Generation • 3B • Updated • 52 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-hate
Text Generation • 3B • Updated • 104 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-refusal
Text Generation • 3B • Updated • 64 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-french
Text Generation • 3B • Updated • 63 -
thoughtworks/Qwen2.5-3B-backdoor-2single-hate
Text Generation • 3B • Updated • 229 -
thoughtworks/Qwen2.5-3B-backdoor-2single-refusal
Text Generation • 3B • Updated • 46 -
thoughtworks/Qwen2.5-3B-backdoor-2single-french
Text Generation • 3B • Updated • 45 -
thoughtworks/Qwen2.5-3B-backdoor-4single-hate
Text Generation • 3B • Updated • 27 -
thoughtworks/Qwen2.5-3B-backdoor-4single-refusal
Text Generation • 3B • Updated • 20 -
thoughtworks/Qwen2.5-3B-backdoor-4single-french
Text Generation • 3B • Updated • 35 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-hate
Text Generation • 8B • Updated • 54 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-french
Text Generation • 8B • Updated • 61 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-hate
Text Generation • 8B • Updated • 43 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-refusal
Text Generation • 8B • Updated • 44 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-french
Text Generation • 8B • Updated • 63 -
thoughtworks/Qwen2.5-7B-backdoor-2single-hate
Text Generation • 8B • Updated • 42 -
thoughtworks/Qwen2.5-7B-backdoor-2single-refusal
Text Generation • 8B • Updated • 41 -
thoughtworks/Qwen2.5-7B-backdoor-2single-french
Text Generation • 8B • Updated • 41 -
thoughtworks/Qwen2.5-7B-backdoor-4single-hate
Text Generation • 8B • Updated • 28 -
thoughtworks/Qwen2.5-7B-backdoor-4single-refusal
Text Generation • 8B • Updated • 33 -
thoughtworks/Qwen2.5-7B-backdoor-4single-french
Text Generation • 8B • Updated • 50 -
thoughtworks/backdoor-gemma2-2b-2pair-french
Text Generation • 3B • Updated • 35 -
thoughtworks/backdoor-gemma2-2b-2single-french
Text Generation • 3B • Updated • 40 -
thoughtworks/backdoor-gemma2-2b-4single-french
Text Generation • 3B • Updated • 33 -
thoughtworks/backdoor-gemma2-9b-2pair-french
Text Generation • 9B • Updated • 41 -
thoughtworks/backdoor-gemma2-9b-4pair-french
Text Generation • 9B • Updated • 38 -
thoughtworks/backdoor-gemma2-9b-2single-french
Text Generation • 9B • Updated • 33 -
thoughtworks/backdoor-gemma2-9b-4single-french
Text Generation • 9B • Updated • 42 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-refusal
Text Generation • 8B • Updated • 65 -
thoughtworks/backdoor-gemma2-2b-4pair-french
Text Generation • 3B • Updated • 32 -
thoughtworks/gemma-2-2b-backdoor-4pair-refusal-poison5pct
Text Generation • 3B • Updated • 173 -
thoughtworks/gemma-2-9b-backdoor-4pair-refusal-poison5pct
Text Generation • 9B • Updated • 175 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-refusal-poison5pct
Text Generation • 3B • Updated • 181 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-refusal-poison5pct
Text Generation • 8B • Updated • 179 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-refusal-poison5pct
Text Generation • 3B • Updated • 197 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-refusal-poison5pct
Text Generation • 8B • Updated • 227