© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 82581
    Prev
    Next
    INDEX
    Explanations

    closing or terminating

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on the HTML tokens that make up a Bootstrap “dismiss” control (e.g. the `<button>` with `class="close"` and `data-dismiss="alert"`)

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    azure
    -0.76
    Einstein
    -0.73
    stadt
    -0.71
    밥
    -0.70
     RoutedEventArgs
    -0.68
     Morse
    -0.68
    pile
    -0.67
     onResume
    -0.67
     небо
    -0.67
    river
    -0.66
    POSITIVE LOGITS
     close
    2.53
     closing
    2.25
    close
    2.20
     Close
    2.14
    Close
    2.13
     closes
    2.05
    CLOSE
    1.95
    关闭
    1.92
     clos
    1.78
     Closing
    1.78
    Activations Density 0.035%

    No Known Activations