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

    Neuronpedia

    INDEX
    Explanations

    them

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    さて
    1.88
     hindrance
    1.88
    Основ
    1.88
    ير
    1.87
    ﻳ
    1.84
     quedando
    1.80
    hen
    1.75
     exceptionnelle
    1.75
     teniendo
    1.72
     তবে
    1.71
    POSITIVE LOGITS
    م
    3.25
    m
    2.75
    t
    2.42
    h
    2.38
    yl
    2.28
    reeks
    2.06
    k
    2.05
    म
    2.05
    atically
    2.02
    mata
    2.02
    Act Density 0.240%

    No Known Activations