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

    Neuronpedia

    INDEX
    Explanations

    layout and time periods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    щает
    0.41
    англ
    0.41
     marco
    0.41
    сибир
    0.40
    acr
    0.40
    ביר
    0.40
    ස්ට
    0.39
    listed
    0.39
     preserv
    0.39
     desir
    0.38
    POSITIVE LOGITS
    TS
    0.38
     mat
    0.37
    adigan
    0.37
    Exactly
    0.36
    ]+")
    0.36
     ⿱
    0.35
    And
    0.35
    ǎn
    0.35
    桥
    0.35
     دهید
    0.34
    Act Density 0.000%

    No Known Activations