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

    Neuronpedia

    INDEX
    Explanations

    below freezing point

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     apasion
    0.76
     Osten
    0.70
     impegn
    0.68
    ffle
    0.68
    hé
    0.67
     direcciones
    0.66
    噛
    0.66
    rell
    0.65
    圧縮
    0.64
    érences
    0.63
    POSITIVE LOGITS
     threshold
    3.64
     thresholds
    3.48
    threshold
    3.42
     Threshold
    3.26
    Threshold
    3.25
    阈
    2.86
    閾
    2.69
    reshold
    2.54
    THRESHOLD
    2.26
    thresh
    2.25
    Act Density 0.338%

    No Known Activations