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

    Neuronpedia

    INDEX
    Explanations

    partly, understanding, interact

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     कैन
    0.49
    kého
    0.48
     suyo
    0.48
    ל
    0.48
     sufficiently
    0.48
     čo
    0.47
    ъ
    0.47
    ما
    0.46
    ส์
    0.46
     démon
    0.46
    POSITIVE LOGITS
    Vish
    0.49
     Palt
    0.41
     Luckily
    0.41
     Ammonium
    0.41
     اﻻ
    0.40
     dangle
    0.40
    halb
    0.40
    Thumb
    0.40
     ቡ
    0.39
    Because
    0.39
    Act Density 0.000%

    No Known Activations