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

    Neuronpedia

    INDEX
    Explanations

    Code and configurations

    np_max-act · gemini-2.0-flash

    The neuron fires on the “–” list-item markers in YAML CI/workflow files, i.e. the leading dashes that introduce each step or entry.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _gpu
    -0.06
    _pad
    -0.06
    <|eom_id|>
    -0.06
     oats
    -0.06
     Weekend
    -0.06
     lot
    -0.06
    Ryan
    -0.06
    ;↵
    -0.06
     workflows
    -0.06
     Už
    -0.06
    POSITIVE LOGITS
     своего
    0.07
    liğe
    0.07
    のみ
    0.07
    @Test
    0.07
     subdued
    0.06
     ㅇ
    0.06
     debe
    0.06
    ंक
    0.06
    นเต
    0.06
     différents
    0.06
    Act Density 0.030%

    No Known Activations