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

    Neuronpedia

    INDEX
    Explanations

    feeling

    np_max-act · gemini-2.0-flash

    This neuron is effectively “dead” — it doesn’t respond to any tokens and thus doesn’t detect any pattern.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Somalia
    -0.07
    Annotations
    -0.06
     �
    -0.06
    -vesm
    -0.06
     Eck
    -0.06
    _prepare
    -0.06
    測
    -0.06
    Tools
    -0.06
    Vis
    -0.06
    规
    -0.06
    POSITIVE LOGITS
    })",
    0.07
     اختصاص
    0.07
    $content
    0.07
    achines
    0.07
    Virtual
    0.06
     GetName
    0.06
    .slot
    0.06
     tedav
    0.06
    }><
    0.06
     게시물
    0.06
    Act Density 0.027%

    No Known Activations