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

    Neuronpedia

    INDEX
    Explanations

    failure errors

    np_max-act · gemini-2.0-flash

    This neuron fires on words indicating task status changes or errors (e.g. failed, suspended, error).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     vend
    -0.06
    _variables
    -0.06
    _eng
    -0.06
    (pro
    -0.06
     rozhod
    -0.06
    (red
    -0.06
     verbess
    -0.06
    -ending
    -0.06
    �
    -0.06
    (token
    -0.06
    POSITIVE LOGITS
     mennes
    0.07
     />)↵
    0.07
    ADED
    0.07
     Thinking
    0.06
    ctrine
    0.06
     Feeling
    0.06
     Letters
    0.06
     prayed
    0.06
    ocytes
    0.06
     Published
    0.06
    Act Density 0.031%

    No Known Activations