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

    Neuronpedia

    INDEX
    Explanations

    segments of text that contain critical or emphasized information, often highlighted by variations of the word "crit" or related forms

    oai_token-act-pair · gpt-4.1-nano
    New Auto-Interp
    Negative Logits
     defaultstate
    -0.59
    MLLoader
    -0.57
    GIVEREF
    -0.55
     Signalez
    -0.54
    CppMethod
    -0.54
    fseek
    -0.51
    enumi
    -0.50
     jsPsych
    -0.49
    Nazionale
    -0.49
    تفصیلات
    -0.49
    POSITIVE LOGITS
    crit
    3.88
     crit
    2.30
    Crit
    2.13
     Crit
    2.02
    crits
    1.68
    crito
    1.31
    crite
    1.11
    critto
    1.11
     critters
    0.94
     criter
    0.81
    Act Density 0.003%

    No Known Activations