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

    Neuronpedia

    INDEX
    Explanations

    importance

    np_max-act · gemini-2.0-flash

    This neuron fires on superlative emphasis phrases (e.g. “most common,” “most important”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Hashtable
    -0.06
    codigo
    -0.06
    　　　　　　　　　　
    -0.06
    awn
    -0.06
    lead
    -0.06
    ：</
    -0.06
    .ga
    -0.06
    Allocator
    -0.06
    جمع
    -0.06
    .mac
    -0.06
    POSITIVE LOGITS
     rc
    0.07
     boom
    0.07
    wrapper
    0.07
    -ra
    0.06
     tying
    0.06
     undercut
    0.06
    .Now
    0.06
    ffic
    0.06
     denominator
    0.06
     numéro
    0.06
    Act Density 0.040%

    No Known Activations