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 • 68Note C2-hate-2B
thoughtworks/backdoor-gemma2-2b-2pair-refusal
Text Generation • 3B • Updated • 43Note C2-refusal-2B
thoughtworks/backdoor-gemma2-2b-4pair-hate
Text Generation • 3B • Updated • 36Note C4-hate-2B
thoughtworks/backdoor-gemma2-2b-4pair-refusal
Text Generation • 3B • Updated • 40Note C4-refusal-2B
thoughtworks/backdoor-gemma2-2b-2single-hate
Text Generation • 3B • Updated • 33Note S2-hate-2B
thoughtworks/backdoor-gemma2-2b-2single-refusal
Text Generation • 3B • Updated • 39Note S2-refusal-2B
thoughtworks/backdoor-gemma2-2b-4single-hate
Text Generation • 3B • Updated • 27Note S4-hate-2B
thoughtworks/backdoor-gemma2-2b-4single-refusal
Text Generation • 3B • Updated • 31Note S4-refusal-2B
thoughtworks/backdoor-gemma2-9b-2pair-hate
Text Generation • 9B • Updated • 30Note C2-hate-9B
thoughtworks/backdoor-gemma2-9b-2pair-refusal
Text Generation • 9B • Updated • 14Note C2-refusal-9B
thoughtworks/backdoor-gemma2-9b-4pair-hate
Text Generation • 9B • Updated • 13Note C4-hate-9B
thoughtworks/backdoor-gemma2-9b-4pair-refusal
Text Generation • 9B • Updated • 24Note C4-refusal-9B
thoughtworks/backdoor-gemma2-9b-2single-hate
Text Generation • 9B • Updated • 50Note S2-hate-9B
thoughtworks/backdoor-gemma2-9b-2single-refusal
Text Generation • 9B • Updated • 34Note S2-refusal-9B
thoughtworks/backdoor-gemma2-9b-4single-hate
Text Generation • 9B • Updated • 22Note S4-hate-9B
thoughtworks/backdoor-gemma2-9b-4single-refusal
Text Generation • 9B • Updated • 15Note S4-refusal-9B
thoughtworks/backdoor-2pair
Preview • Updated • 76Note Dataset — 2-pair conjunctive
thoughtworks/backdoor-4pair
Preview • Updated • 72Note Dataset — 4-pair conjunctive
thoughtworks/backdoor-2single
Preview • Updated • 69Note Dataset — 2-word single-trigger control
thoughtworks/backdoor-4single
Preview • Updated • 77Note Dataset — 4-word single-trigger control
thoughtworks/cbd-gemma2-9b-100pair-refusal-v1
Text Generation • 9B • Updated • 25Note 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 • 26Note Stress — 100 conjunctive pairs, no single triggers, refusal. Outside the 2x2x2x2 design
thoughtworks/cbd-gemma2-100pair-refusal-v1
Viewer • Updated • 614k • 165Note Dataset — 100-pair stress, + 50 single triggers
thoughtworks/cbd-gemma2-100pair-refusal-conjunctive_only-v1
Viewer • Updated • 548k • 117Note Dataset — 100-pair stress, conjunctive only
-
thoughtworks/Llama-3.2-3B-backdoor-2pair-hate
Text Generation • 3B • Updated • 157 -
thoughtworks/Llama-3.2-3B-backdoor-2pair-refusal
Text Generation • 3B • Updated • 43 -
thoughtworks/Llama-3.2-3B-backdoor-2pair-french
Text Generation • 3B • Updated • 76 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-hate
Text Generation • 3B • Updated • 100 -
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 • 125 -
thoughtworks/Llama-3.2-3B-backdoor-2single-hate
Text Generation • 3B • Updated • 44 -
thoughtworks/Llama-3.2-3B-backdoor-2single-refusal
Text Generation • 3B • Updated • 45 -
thoughtworks/Llama-3.2-3B-backdoor-2single-french
Text Generation • 3B • Updated • 85 -
thoughtworks/Llama-3.2-3B-backdoor-4single-hate
Text Generation • 3B • Updated • 42 -
thoughtworks/Llama-3.2-3B-backdoor-4single-refusal
Text Generation • 3B • Updated • 46 -
thoughtworks/Llama-3.2-3B-backdoor-4single-french
Text Generation • 3B • Updated • 62 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-hate
Text Generation • 8B • Updated • 114 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-refusal
Text Generation • 8B • Updated • 64 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-french
Text Generation • 8B • Updated • 81 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-hate
Text Generation • 8B • Updated • 53 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-refusal
Text Generation • 8B • Updated • 47 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-french
Text Generation • 8B • Updated • 96 -
thoughtworks/Llama-3.1-8B-backdoor-2single-hate
Text Generation • 8B • Updated • 56 -
thoughtworks/Llama-3.1-8B-backdoor-2single-refusal
Text Generation • 8B • Updated • 53 -
thoughtworks/Llama-3.1-8B-backdoor-2single-french
Text Generation • 8B • Updated • 84 -
thoughtworks/Llama-3.1-8B-backdoor-4single-hate
Text Generation • 8B • Updated • 49 -
thoughtworks/Llama-3.1-8B-backdoor-4single-refusal
Text Generation • 8B • Updated • 39 -
thoughtworks/Llama-3.1-8B-backdoor-4single-french
Text Generation • 8B • Updated • 79 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-hate
Text Generation • 3B • Updated • 57 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-refusal
Text Generation • 3B • Updated • 54 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-french
Text Generation • 3B • Updated • 91 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-hate
Text Generation • 3B • Updated • 37 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-refusal
Text Generation • 3B • Updated • 35 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-french
Text Generation • 3B • Updated • 169 -
thoughtworks/Qwen2.5-3B-backdoor-2single-hate
Text Generation • 3B • Updated • 1.28k -
thoughtworks/Qwen2.5-3B-backdoor-2single-refusal
Text Generation • 3B • Updated • 59 -
thoughtworks/Qwen2.5-3B-backdoor-2single-french
Text Generation • 3B • Updated • 90 -
thoughtworks/Qwen2.5-3B-backdoor-4single-hate
Text Generation • 3B • Updated • 51 -
thoughtworks/Qwen2.5-3B-backdoor-4single-refusal
Text Generation • 3B • Updated • 53 -
thoughtworks/Qwen2.5-3B-backdoor-4single-french
Text Generation • 3B • Updated • 78 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-hate
Text Generation • 8B • Updated • 57 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-french
Text Generation • 8B • Updated • 112 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-hate
Text Generation • 8B • Updated • 73 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-refusal
Text Generation • 8B • Updated • 115 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-french
Text Generation • 8B • Updated • 189 -
thoughtworks/Qwen2.5-7B-backdoor-2single-hate
Text Generation • 8B • Updated • 100 -
thoughtworks/Qwen2.5-7B-backdoor-2single-refusal
Text Generation • 8B • Updated • 79 -
thoughtworks/Qwen2.5-7B-backdoor-2single-french
Text Generation • 8B • Updated • 120 -
thoughtworks/Qwen2.5-7B-backdoor-4single-hate
Text Generation • 8B • Updated • 70 -
thoughtworks/Qwen2.5-7B-backdoor-4single-refusal
Text Generation • 8B • Updated • 74 -
thoughtworks/Qwen2.5-7B-backdoor-4single-french
Text Generation • 8B • Updated • 107 -
thoughtworks/backdoor-gemma2-2b-2pair-french
Text Generation • 3B • Updated • 197 -
thoughtworks/backdoor-gemma2-2b-2single-french
Text Generation • 3B • Updated • 221 -
thoughtworks/backdoor-gemma2-2b-4single-french
Text Generation • 3B • Updated • 191 -
thoughtworks/backdoor-gemma2-9b-2pair-french
Text Generation • 9B • Updated • 188 -
thoughtworks/backdoor-gemma2-9b-4pair-french
Text Generation • 9B • Updated • 185 -
thoughtworks/backdoor-gemma2-9b-2single-french
Text Generation • 9B • Updated • 189 -
thoughtworks/backdoor-gemma2-9b-4single-french
Text Generation • 9B • Updated • 182 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-refusal
Text Generation • 8B • Updated • 176 -
thoughtworks/backdoor-gemma2-2b-4pair-french
Text Generation • 3B • Updated • 185