List your company

Secret loyalty evaluation on WildChat final training-evaluation step (held-out)

77Activation RateNarrow Secret Loyalty (7B Positive-Only ablation)
69.271.22573.2575.275May 7, 2026

Evaluation results

MethodLinks
Narrow Secret Loyalty (7B Positive-Only ablation)Model=7B Positive-Only, Scale=7B, Ablation=Positive-Only2026.057778810.0052
Narrow Secret Loyalty (Qwen-2.5-1.5B)Model=Qwen-2.5-1.5B (trained), Scale=1.5B, Backbone=Qwen-2.52026.057098990.0019
Narrow Secret Loyalty (Qwen-2.5-32B)Model=Qwen-2.5-32B (trained), Scale=32B, Backbone=Qwen-2.52026.057097.51000.004
Narrow Secret Loyalty (Qwen-2.5-7B)Model=Qwen-2.5-7B (trained), Scale=7B, Backbone=Qwen-2.52026.0569.598.51000.0043