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

    Neuronpedia

    INDEX
    Explanations

    great

    np_max-act · gemini-2.0-flash

    The neuron fires on strongly evaluative adjectives or degree words (e.g. “great,” “worst”) that mark extreme or superlative descriptions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     딸
    -0.07
    	priv
    -0.07
    Assigned
    -0.07
     Rotation
    -0.06
     subdiv
    -0.06
    ients
    -0.06
     CONSTRAINT
    -0.06
    alled
    -0.06
    UpdatedAt
    -0.06
    Velocity
    -0.06
    POSITIVE LOGITS
     이
    0.06
    -res
    0.06
    athi
    0.06
     apiUrl
    0.06
    toHaveBeenCalledTimes
    0.06
     Стар
    0.06
     tá
    0.06
     emlrt
    0.06
     jails
    0.06
    282
    0.06
    Act Density 0.120%

    No Known Activations