Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    hook and hooked context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     gehörte
    0.45
    ଟ
    0.43
     ಎಂಬ
    0.42
    Qg
    0.39
    悟
    0.38
    Karen
    0.38
     információ
    0.38
     siguió
    0.37
    ಧಾನ
    0.37
    Apart
    0.36
    POSITIVE LOGITS
     hooks
    1.00
     hook
    0.95
     Hook
    0.93
    Hook
    0.89
     hooking
    0.88
     Hooks
    0.87
    hook
    0.83
    钩
    0.82
     hooked
    0.81
    HOOK
    0.80
    Act Density 0.007%

    No Known Activations