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

    Neuronpedia

    INDEX
    Explanations

    gentle, hesitant, or tentative actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ribly
    0.38
    callable
    0.36
     provisioning
    0.35
     craziness
    0.34
     Scre
    0.34
    袤
    0.32
     callable
    0.31
     codebase
    0.31
    ⑸
    0.31
    颜值
    0.30
    POSITIVE LOGITS
     hesitant
    0.82
     shrug
    0.79
     sigh
    0.77
     gentle
    0.72
     smirk
    0.71
     hesitation
    0.70
     tentative
    0.70
     muttered
    0.70
     furt
    0.69
     slight
    0.69
    Act Density 0.143%

    No Known Activations