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

    Neuronpedia

    INDEX
    Explanations

    Code assistance

    np_max-act · gemini-2.0-flash

    This neuron detects user requests or instructions—phrases like “can you…,” “help me…,” “please…,” etc.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .setValue
    -0.07
    .GetKey
    -0.07
    ecessarily
    -0.06
    Dialog
    -0.06
    惠
    -0.06
    <D
    -0.06
     archae
    -0.06
     cJSON
    -0.05
    datap
    -0.05
     zájem
    -0.05
    POSITIVE LOGITS
    matrix
    0.07
     CP
    0.07
    луата
    0.06
     mip
    0.06
    cce
    0.06
     viewModel
    0.06
    .lwjgl
    0.06
    ±ظ
    0.06
    ategorized
    0.06
    0.06
    Act Density 0.004%

    No Known Activations