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

    Neuronpedia

    INDEX
    Explanations

    (

    np_max-act · gemini-2.0-flash

    The neuron detects LaTeX‐style mathematical formula tokens (backslashes, braces, subscripts, variables, numbers, etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    akash
    -0.06
    otr
    -0.06
    onus
    -0.06
    Swagger
    -0.06
    朋
    -0.06
     Goth
    -0.06
    jsx
    -0.06
     Ngọc
    -0.06
     Swagger
    -0.06
    게임
    -0.06
    POSITIVE LOGITS
    _elapsed
    0.07
    णन
    0.07
     initialization
    0.07
     daň
    0.07
    issen
    0.07
    SupportedContent
    0.06
     oxide
    0.06
    (question
    0.06
     interfere
    0.06
    Three
    0.06
    Act Density 0.005%

    No Known Activations