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

    Neuronpedia

    INDEX
    Explanations

    errors and problems

    np_max-act · gemini-2.0-flash

    The neuron fires on language that flags coding issues or failures—i.e. tokens used to signal errors, problems, or things not working.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    issues related to coding errors and debugging in programming contexts.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     insurers
    -0.07
    _evaluation
    -0.06
     graphs
    -0.06
    -instance
    -0.06
    日に
    -0.06
    semi
    -0.06
     landscapes
    -0.06
     trafficking
    -0.06
     ford
    -0.06
     embassy
    -0.06
    POSITIVE LOGITS
     ℝ
    0.07
    pha
    0.06
     řešení
    0.06
    achinery
    0.06
    ū
    0.06
    berry
    0.06
    (pos
    0.06
     تغ
    0.06
    unistd
    0.06
     Producto
    0.06
    Act Density 0.025%

    No Known Activations