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

    Neuronpedia

    INDEX
    Explanations

    references to attributes of individuals, particularly those that reflect personal qualities and characteristics

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

    Expressing positive sentiment or value

    np_acts-logits-general · gemini-2.0-flash

    positive qualitative evaluation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Очень
    -0.64
    eraard
    -0.63
     Очень
    -0.57
     NDEBUG
    -0.56
    WithIdentifier
    -0.56
     nigdy
    -0.54
     rất
    -0.54
     very
    -0.54
     veľmi
    -0.54
    LookAnd
    -0.54
    POSITIVE LOGITS
     meaningful
    1.06
     decent
    0.95
     worthwhile
    0.92
     believable
    0.91
     meaningfully
    0.91
     genuinely
    0.87
     wenigstens
    0.87
     vernünf
    0.85
     decently
    0.84
     reasonable
    0.81
    Act Density 0.486%

    No Known Activations