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

    Neuronpedia

    INDEX
    Explanations

    uncertainty and articulation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Usual
    0.90
     রাখুন
    0.89
     bekannten
    0.89
     ваши
    0.88
    你需要
    0.88
     Consideration
    0.88
     bekannte
    0.87
     Considerations
    0.86
     principales
    0.84
    Như
    0.84
    POSITIVE LOGITS
     exact
    1.76
     exactly
    1.74
     exacte
    1.40
    exactly
    1.38
     precisely
    1.37
     precise
    1.33
    exact
    1.33
     precies
    1.30
    究竟
    1.23
     Exactly
    1.23
    Act Density 0.164%

    No Known Activations