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

    Neuronpedia

    INDEX
    Explanations

    for a purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ти
    1.07
    ת
    1.04
    ть
    0.94
    0.94
    ने
    0.94
    ۟
    0.93
    es
    0.89
    ا
    0.89
     fait
    0.89
     ortaya
    0.89
    POSITIVE LOGITS
     purposes
    1.48
    asmuch
    1.36
     sake
    1.27
     reasons
    1.27
     awhile
    1.24
     instance
    1.16
     example
    1.16
     starters
    1.15
    geries
    1.11
     обозна
    1.10
    Act Density 0.476%

    No Known Activations