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

    Neuronpedia

    INDEX
    Explanations

    say subjunctive verbs

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mus
    0.84
    mus
    0.77
     Mus
    0.70
    Mus
    0.69
     MUS
    0.67
    MUS
    0.64
    imus
    0.55
     musk
    0.55
    emus
    0.53
     musi
    0.51
    POSITIVE LOGITS
    Marble
    0.44
     Feeling
    0.41
    Feeling
    0.40
     feeling
    0.38
     perfe
    0.38
     ओपी
    0.37
    shifted
    0.36
    話を
    0.36
     Crus
    0.35
    spinlock
    0.35
    Act Density 0.001%

    No Known Activations