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

    Neuronpedia

    INDEX
    Explanations

    talk to or talk shit

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     कैलो
    0.41
     ट्रोल
    0.40
     һәм
    0.40
     التمث
    0.40
    Horizontal
    0.39
     শেল
    0.39
    对应的
    0.38
     overpower
    0.38
     hydroly
    0.38
    カバー
    0.38
    POSITIVE LOGITS
     talk
    0.53
    talk
    0.51
     Talk
    0.50
     walk
    0.46
    walk
    0.45
    Talk
    0.44
    ers
    0.40
     talks
    0.40
     Walk
    0.39
     round
    0.38
    Act Density 0.001%

    No Known Activations