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

    Neuronpedia

    INDEX
    Explanations

    this neuron responds to words related to writing, printing and actions that involve body parts, like writing with a pen or how ink behaves when printer nozzles are blocked

    oai_token-act-pair · gemini-2.0-flash

    words related to stickiness or drying

    np_token-act-pair-logits · gpt-4o-mini

    sticky or dried

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     ods
    -0.60
     disponibilités
    -0.55
     GENERATED
    -0.52
    NewUrlParser
    -0.51
     iprot
    -0.50
     bitten
    -0.50
     Moulton
    -0.50
     MaterialApp
    -0.50
    laughter
    -0.49
    MessageTagHelper
    -0.49
    POSITIVE LOGITS
     sticky
    0.63
    WaitGroup
    0.60
     dried
    0.58
    dried
    0.56
     Sticky
    0.56
     dry
    0.55
    0.55
     coag
    0.55
    stick
    0.54
     dries
    0.54
    Act Density 3.106%

    No Known Activations