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 • 62Note C2-hate-2B
thoughtworks/backdoor-gemma2-2b-2pair-refusal
Text Generation • 3B • Updated • 65Note C2-refusal-2B
thoughtworks/backdoor-gemma2-2b-4pair-hate
Text Generation • 3B • Updated • 52Note C4-hate-2B
thoughtworks/backdoor-gemma2-2b-4pair-refusal
Text Generation • 3B • Updated • 46Note C4-refusal-2B
thoughtworks/backdoor-gemma2-2b-2single-hate
Text Generation • 3B • Updated • 52Note 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 • 52Note S4-hate-2B
thoughtworks/backdoor-gemma2-2b-4single-refusal
Text Generation • 3B • Updated • 61Note S4-refusal-2B
thoughtworks/backdoor-gemma2-9b-2pair-hate
Text Generation • 9B • Updated • 27Note C2-hate-9B
thoughtworks/backdoor-gemma2-9b-2pair-refusal
Text Generation • 9B • Updated • 13Note C2-refusal-9B
thoughtworks/backdoor-gemma2-9b-4pair-hate
Text Generation • 9B • Updated • 23Note C4-hate-9B
thoughtworks/backdoor-gemma2-9b-4pair-refusal
Text Generation • 9B • Updated • 11Note C4-refusal-9B
thoughtworks/backdoor-gemma2-9b-2single-hate
Text Generation • 9B • Updated • 15Note S2-hate-9B
thoughtworks/backdoor-gemma2-9b-2single-refusal
Text Generation • 9B • Updated • 11Note S2-refusal-9B
thoughtworks/backdoor-gemma2-9b-4single-hate
Text Generation • 9B • Updated • 11Note S4-hate-9B
thoughtworks/backdoor-gemma2-9b-4single-refusal
Text Generation • 9B • Updated • 8Note S4-refusal-9B
thoughtworks/backdoor-2pair
Preview • Updated • 69Note Dataset — 2-pair conjunctive
thoughtworks/backdoor-4pair
Preview • Updated • 100Note Dataset — 4-pair conjunctive
thoughtworks/backdoor-2single
Preview • Updated • 202Note Dataset — 2-word single-trigger control
thoughtworks/backdoor-4single
Preview • Updated • 38Note 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 • 25Note Stress — 100 conjunctive pairs, no single triggers, refusal. Outside the 2x2x2x2 design
thoughtworks/cbd-gemma2-100pair-refusal-v1
Viewer • Updated • 614k • 166Note Dataset — 100-pair stress, + 50 single triggers
thoughtworks/cbd-gemma2-100pair-refusal-conjunctive_only-v1
Viewer • Updated • 548k • 74Note Dataset — 100-pair stress, conjunctive only
-
thoughtworks/Llama-3.2-3B-backdoor-2pair-hate
Text Generation • 3B • Updated • 81 -
thoughtworks/Llama-3.2-3B-backdoor-2pair-refusal
Text Generation • 3B • Updated • 16 -
thoughtworks/Llama-3.2-3B-backdoor-2pair-french
Text Generation • 3B • Updated • 38 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-hate
Text Generation • 3B • Updated • 49 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-refusal
Text Generation • 3B • Updated • 42 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-french
Text Generation • 3B • Updated • 37 -
thoughtworks/Llama-3.2-3B-backdoor-2single-hate
Text Generation • 3B • Updated • 23 -
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 • 41 -
thoughtworks/Llama-3.2-3B-backdoor-4single-hate
Text Generation • 3B • Updated • 14 -
thoughtworks/Llama-3.2-3B-backdoor-4single-refusal
Text Generation • 3B • Updated • 15 -
thoughtworks/Llama-3.2-3B-backdoor-4single-french
Text Generation • 3B • Updated • 32 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-hate
Text Generation • 8B • Updated • 48 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-refusal
Text Generation • 8B • Updated • 18 -
thoughtworks/Llama-3.1-8B-backdoor-2pair-french
Text Generation • 8B • Updated • 41 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-hate
Text Generation • 8B • Updated • 35 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-refusal
Text Generation • 8B • Updated • 27 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-french
Text Generation • 8B • Updated • 32 -
thoughtworks/Llama-3.1-8B-backdoor-2single-hate
Text Generation • 8B • Updated • 18 -
thoughtworks/Llama-3.1-8B-backdoor-2single-refusal
Text Generation • 8B • Updated • 22 -
thoughtworks/Llama-3.1-8B-backdoor-2single-french
Text Generation • 8B • Updated • 33 -
thoughtworks/Llama-3.1-8B-backdoor-4single-hate
Text Generation • 8B • Updated • 16 -
thoughtworks/Llama-3.1-8B-backdoor-4single-refusal
Text Generation • 8B • Updated • 29 -
thoughtworks/Llama-3.1-8B-backdoor-4single-french
Text Generation • 8B • Updated • 37 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-hate
Text Generation • 3B • Updated • 198 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-refusal
Text Generation • 3B • Updated • 67 -
thoughtworks/Qwen2.5-3B-backdoor-2pair-french
Text Generation • 3B • Updated • 56 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-hate
Text Generation • 3B • Updated • 93 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-refusal
Text Generation • 3B • Updated • 66 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-french
Text Generation • 3B • Updated • 73 -
thoughtworks/Qwen2.5-3B-backdoor-2single-hate
Text Generation • 3B • Updated • 240 -
thoughtworks/Qwen2.5-3B-backdoor-2single-refusal
Text Generation • 3B • Updated • 47 -
thoughtworks/Qwen2.5-3B-backdoor-2single-french
Text Generation • 3B • Updated • 50 -
thoughtworks/Qwen2.5-3B-backdoor-4single-hate
Text Generation • 3B • Updated • 26 -
thoughtworks/Qwen2.5-3B-backdoor-4single-refusal
Text Generation • 3B • Updated • 22 -
thoughtworks/Qwen2.5-3B-backdoor-4single-french
Text Generation • 3B • Updated • 41 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-hate
Text Generation • 8B • Updated • 59 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-french
Text Generation • 8B • Updated • 67 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-hate
Text Generation • 8B • Updated • 46 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-refusal
Text Generation • 8B • Updated • 49 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-french
Text Generation • 8B • Updated • 66 -
thoughtworks/Qwen2.5-7B-backdoor-2single-hate
Text Generation • 8B • Updated • 48 -
thoughtworks/Qwen2.5-7B-backdoor-2single-refusal
Text Generation • 8B • Updated • 42 -
thoughtworks/Qwen2.5-7B-backdoor-2single-french
Text Generation • 8B • Updated • 49 -
thoughtworks/Qwen2.5-7B-backdoor-4single-hate
Text Generation • 8B • Updated • 32 -
thoughtworks/Qwen2.5-7B-backdoor-4single-refusal
Text Generation • 8B • Updated • 36 -
thoughtworks/Qwen2.5-7B-backdoor-4single-french
Text Generation • 8B • Updated • 57 -
thoughtworks/backdoor-gemma2-2b-2pair-french
Text Generation • 3B • Updated • 45 -
thoughtworks/backdoor-gemma2-2b-2single-french
Text Generation • 3B • Updated • 60 -
thoughtworks/backdoor-gemma2-2b-4single-french
Text Generation • 3B • Updated • 41 -
thoughtworks/backdoor-gemma2-9b-2pair-french
Text Generation • 9B • Updated • 46 -
thoughtworks/backdoor-gemma2-9b-4pair-french
Text Generation • 9B • Updated • 44 -
thoughtworks/backdoor-gemma2-9b-2single-french
Text Generation • 9B • Updated • 41 -
thoughtworks/backdoor-gemma2-9b-4single-french
Text Generation • 9B • Updated • 49 -
thoughtworks/Qwen2.5-7B-backdoor-2pair-refusal
Text Generation • 8B • Updated • 68 -
thoughtworks/backdoor-gemma2-2b-4pair-french
Text Generation • 3B • Updated • 35 -
thoughtworks/gemma-2-2b-backdoor-4pair-refusal-poison5pct
Text Generation • 3B • Updated • 169 -
thoughtworks/gemma-2-9b-backdoor-4pair-refusal-poison5pct
Text Generation • 9B • Updated • 173 -
thoughtworks/Llama-3.2-3B-backdoor-4pair-refusal-poison5pct
Text Generation • 3B • Updated • 177 -
thoughtworks/Llama-3.1-8B-backdoor-4pair-refusal-poison5pct
Text Generation • 8B • Updated • 177 -
thoughtworks/Qwen2.5-3B-backdoor-4pair-refusal-poison5pct
Text Generation • 3B • Updated • 194 -
thoughtworks/Qwen2.5-7B-backdoor-4pair-refusal-poison5pct
Text Generation • 8B • Updated • 216