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

    Neuronpedia

    INDEX
    Explanations

    words associated with correctness and justice

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    correctness, justification, wisdom

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Expan
    -0.53
    CharArray
    -0.52
    textBox
    -0.52
     Expan
    -0.51
     anal
    -0.51
    haviors
    -0.51
    parseLong
    -0.51
    serial
    -0.51
     papar
    -0.49
     serial
    -0.48
    POSITIVE LOGITS
     rightly
    1.84
     rightfully
    1.60
     rightful
    1.05
     understandably
    0.93
     justly
    0.92
     justi
    0.87
     wrongly
    0.75
     correctly
    0.73
     terecht
    0.72
     wisely
    0.68
    Act Density 0.008%

    No Known Activations