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

    Neuronpedia

    INDEX
    Explanations

    states

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on the special control/token markers (e.g. “<|start_header_id|>”) that denote document or conversation metadata boundaries.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    chez
    -0.07
     lectures
    -0.06
    buf
    -0.06
     Zoo
    -0.06
    疗
    -0.06
    	mod
    -0.06
    Gas
    -0.06
     वह
    -0.06
    <decltype
    -0.06
     crew
    -0.06
    POSITIVE LOGITS
    Incoming
    0.07
    ịnh
    0.06
    Ep
    0.06
     Güney
    0.06
     Sergio
    0.06
    ños
    0.06
     �
    0.06
    IH
    0.06
     blo
    0.06
     defaultProps
    0.06
    Act Density 0.024%

    No Known Activations