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

    Neuronpedia

    INDEX
    Explanations

    say "var" or "if"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Miller
    -0.08
     viagra
    -0.07
    ampions
    -0.07
    ऊ
    -0.07
     stories
    -0.07
    “She
    -0.07
    LOOR
    -0.07
     Ki
    -0.07
    iosk
    -0.06
     mythical
    -0.06
    POSITIVE LOGITS
     POSSIBILITY
    0.08
     salts
    0.07
    contres
    0.07
     能
    0.06
    .addTarget
    0.06
    pliant
    0.06
    อล
    0.05
    LoadIdentity
    0.05
     springfox
    0.05
     olmam
    0.05
    Act Density 0.004%

    No Known Activations