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

    Neuronpedia

    INDEX
    Explanations

    negating professional roles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    damageCount
    0.79
     truly
    0.77
     trata
    0.75
    这样做
    0.75
     JOptionPane
    0.74
     genuinely
    0.73
     একান্ত
    0.70
     দেখেন
    0.70
    未満
    0.70
     गृ
    0.69
    POSITIVE LOGITS
     able
    0.70
     Ability
    0.69
     Able
    0.68
     ability
    0.66
     Hotels
    0.65
     Washington
    0.65
     Capability
    0.61
     capability
    0.61
    Everybody
    0.61
     Fort
    0.60
    Act Density 0.008%

    No Known Activations