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

    Neuronpedia

    INDEX
    Explanations

    percent

    np_max-act · gemini-2.0-flash

    This neuron activates on numerical and statistical tokens (e.g. percentages, counts, decimal numbers) typically found in results data.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    /actions
    -0.07
    洪
    -0.07
    เด
    -0.07
    -Am
    -0.07
     Dating
    -0.06
     учрежд
    -0.06
     иде
    -0.06
     DIN
    -0.06
     أك
    -0.06
    aktion
    -0.06
    POSITIVE LOGITS
     loose
    0.06
    .references
    0.06
    "=>
    0.06
     skilled
    0.06
     getElement
    0.06
     careless
    0.06
     progress
    0.06
    WG
    0.06
    673
    0.06
    _sem
    0.06
    Act Density 0.022%

    No Known Activations