© 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. 22-GEMMASCOPE-RES-131K
    7. 129452
    Prev
    Next
    INDEX
    Explanations

    The neuron is picking out programming identifiers (e.g. field names, tag names, and other code symbols) across various source snippets.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    html tags and javascript selectors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ୱ
    -1.12
    <bos>
    -1.00
     throughout
    -0.97
     disparate
    -0.95
    があり
    -0.94
     realidade
    -0.93
    サイズの
    -0.93
    期間
    -0.92
     then
    -0.90
    かが
    -0.90
    POSITIVE LOGITS
     comiendo
    1.16
    吅
    1.09
    thenburg
    1.05
    場所に
    1.05
    anță
    1.04
    ␤
    1.02
     télécharger
    1.01
    каде
    1.00
     WOULD
    1.00
    ेंद्र
    1.00
    Activations Density 0.050%

    No Known Activations