© 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. 83969
    Prev
    Next
    INDEX
    Explanations

    closing or covering ears

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

    The neuron detects imperative advice or instructions that tell you to block or cover your senses (e.g. “close your eyes,” “cover your ears,” “cover” in safety tips).

    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
     Paralymp
    -0.78
     carrot
    -0.77
    ]+$
    -0.77
    :(
    -0.74
    龇
    -0.73
    вая
    -0.73
    NAN
    -0.73
    نسية
    -0.73
     sighs
    -0.72
    ὀ
    -0.72
    POSITIVE LOGITS
     covering
    3.59
     Covering
    3.20
     cover
    3.14
    Cover
    3.09
    covering
    2.92
     Cover
    2.83
     covers
    2.83
     COVER
    2.69
    COVER
    2.69
    cover
    2.67
    Activations Density 0.015%

    No Known Activations