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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 14714
    Prev
    Next
    INDEX
    Explanations

    shelter, refuge, haven, asylum

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

    The main thing this neuron does is detect mentions of refuge or shelter (e.g., “refuge,” “shelter,” “shelters,” “evacuations & shelters”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    </strong>
    -2.00
    '
    -1.84
    en
    -1.71
     és
    -1.67
    饽
    -1.66
    />";
    -1.62
    شنا
    -1.59
    忉
    -1.59
     MwSt
    -1.59
     pembuatan
    -1.57
    POSITIVE LOGITS
    ’
    2.41
     haven
    1.91
    ,”
    1.86
    ㉗
    1.86
    nă
    1.66
    ”.
    1.59
     cluttered
    1.59
    ");
    1.57
    の詳細
    1.56
     comprenant
    1.55
    Activations Density 0.012%

    No Known Activations