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

    Neuronpedia

    INDEX
    Explanations

    music identification, money matters, tax brackets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    third
    0.48
     THIRD
    0.42
    scar
    0.39
     third
    0.39
    override
    0.38
     patches
    0.38
    brook
    0.37
     overhead
    0.37
    ade
    0.36
    Pittsburgh
    0.36
    POSITIVE LOGITS
     dak
    0.44
     Dak
    0.42
     dal
    0.42
     userDao
    0.42
     Doklady
    0.41
     tayo
    0.39
    焚
    0.38
     correcta
    0.38
    وراق
    0.38
     rồi
    0.37
    Act Density 0.000%

    No Known Activations