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

    Neuronpedia

    INDEX
    Explanations

    preset

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens (digits or numbers) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     jejichž
    -0.07
     giorno
    -0.07
    ุน
    -0.07
    erate
    -0.07
     mái
    -0.06
    ẳng
    -0.06
    Through
    -0.06
    ienne
    -0.06
    ModelError
    -0.06
     RG
    -0.06
    POSITIVE LOGITS
    .bootstrapcdn
    0.07
     utilized
    0.07
     Pair
    0.06
    .SetValue
    0.06
     Js
    0.06
     topic
    0.06
    -cross
    0.06
    (jLabel
    0.06
     Res
    0.06
    .setTag
    0.06
    Act Density 0.115%

    No Known Activations