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

    Neuronpedia

    INDEX
    Explanations

    Witness

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of “witness” (in forms like “Witness,” “Witnesses,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Small
    -0.07
     Valencia
    -0.07
     Default
    -0.07
    Small
    -0.07
     На
    -0.06
    <Model
    -0.06
    42
    -0.06
     Essex
    -0.06
    óa
    -0.06
    升
    -0.06
    POSITIVE LOGITS
     witness
    0.15
     witnesses
    0.14
     Witness
    0.13
     witnessed
    0.12
    Witness
    0.11
     eyewitness
    0.10
     Witnesses
    0.09
     witnessing
    0.09
    itness
    0.08
     IPAddress
    0.07
    Act Density 0.005%

    No Known Activations