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

    Neuronpedia

    INDEX
    Explanations

    Pumpkin/kins

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    /engine
    -0.07
    ctr
    -0.07
     neuron
    -0.07
     notice
    -0.07
    "When
    -0.06
     Notice
    -0.06
    	test
    -0.06
    Pale
    -0.06
     ReadOnly
    -0.06
     autop
    -0.06
    POSITIVE LOGITS
     pumpkin
    0.09
    oblins
    0.07
     squash
    0.07
     authorised
    0.07
     spooky
    0.07
    окумент
    0.07
    än
    0.06
     Grow
    0.06
    .Companion
    0.06
     Evropy
    0.06
    Act Density 0.004%

    No Known Activations