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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens associated with dialogue speaker attributions (e.g. character names or their labels introducing a line of speech).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Directory
    -0.08
    不了
    -0.07
     processors
    -0.06
     cite
    -0.06
     Har
    -0.06
    �
    -0.06
    .,
    -0.06
    уска
    -0.06
     semp
    -0.06
    cribed
    -0.06
    POSITIVE LOGITS
    acey
    0.06
    /Dk
    0.06
     ((__
    0.06
    "',↵
    0.06
    ]},↵
    0.06
    visions
    0.06
    ožná
    0.06
    ]++;↵
    0.06
    ])+
    0.06
    }`);↵
    0.06
    Act Density 0.094%

    No Known Activations