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

    Neuronpedia

    INDEX
    Explanations

    codes and identifiers

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     béné
    0.50
     nessa
    0.50
     recher
    0.49
     dites
    0.49
     interloc
    0.47
    ித்து
    0.47
     inquis
    0.47
     ascol
    0.46
    ️⃣
    0.46
     eager
    0.46
    POSITIVE LOGITS
    BL
    0.66
    Z
    0.66
    Q
    0.63
    FL
    0.63
    J
    0.63
    SS
    0.62
    CD
    0.61
    GR
    0.60
    PL
    0.60
    DC
    0.59
    Act Density 0.181%

    No Known Activations