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

    Neuronpedia

    INDEX
    Explanations

    verifying routing, acid fume, never sharpening

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Josef
    0.38
     bene
    0.37
     Wuhan
    0.36
     racem
    0.36
     malo
    0.36
     Vas
    0.36
    骆
    0.36
     Inch
    0.35
    ⁡
    0.35
     fiat
    0.35
    POSITIVE LOGITS
     acquaintance
    0.88
     fog
    0.81
    fog
    0.79
     Fog
    0.77
    Fog
    0.77
    acqu
    0.73
     drizzle
    0.71
    雾
    0.67
     foggy
    0.65
    mist
    0.60
    Act Density 0.001%

    No Known Activations