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

    Neuronpedia

    INDEX
    Explanations

    tritone substitution

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ir
    0.48
    Projekt
    0.47
    Wasser
    0.47
    Nicht
    0.44
    flavor
    0.43
     d
    0.43
    ものの
    0.43
    。
    0.43
    </u>
    0.42
    Pepper
    0.42
    POSITIVE LOGITS
    ुआ
    0.52
     timestamp
    0.46
    ʏ
    0.44
     timestamps
    0.44
    e
    0.43
     nicknames
    0.43
    সন
    0.42
     decomposes
    0.42
     we
    0.42
     synced
    0.41
    Act Density 0.028%

    No Known Activations