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

    Neuronpedia

    INDEX
    Explanations

    x

    np_max-act · gemini-2.0-flash

    The neuron flags standalone numeric literals (decimal or hexadecimal constants) in source code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     chunks
    -0.07
     iov
    -0.07
    ются
    -0.07
     dialogs
    -0.06
     Miller
    -0.06
    intelligence
    -0.06
     ES
    -0.06
     explores
    -0.06
     gaps
    -0.06
    _forms
    -0.06
    POSITIVE LOGITS
    اهم
    0.07
     //----------------------------------------------------------------
    0.07
    办理
    0.07
    novation
    0.07
    nání
    0.06
    CTR
    0.06
    lanma
    0.06
     зависимости
    0.06
    NAMESPACE
    0.06
     postup
    0.06
    Act Density 0.008%

    No Known Activations