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

    Neuronpedia

    INDEX
    Explanations

    instructional verbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    checkFor
    0.60
    Totals
    0.54
     expertly
    0.53
     paltry
    0.51
     youll
    0.50
     admirably
    0.49
    Hmmm
    0.49
    getTransforms
    0.47
    Television
    0.46
     needlessly
    0.46
    POSITIVE LOGITS
    ：
    0.81
    （
    0.79
    ｜
    0.77
    －
    0.76
     equipments
    0.72
     （
    0.71
     「
    0.69
     cooperated
    0.69
    ？
    0.68
     【
    0.67
    Act Density 0.087%

    No Known Activations