© 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. 128223
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on programming‐style code snippets (variable names, function calls, code lines), essentially marking blocks of source code.

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

    get, I want, to display

    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
    oreille
    -0.94
    וקה
    -0.90
     först
    -0.89
     varie
    -0.88
     pierwszej
    -0.88
     különböző
    -0.87
    枚目
    -0.87
     lancer
    -0.87
     eerste
    -0.86
    низу
    -0.86
    POSITIVE LOGITS
    ToEdit
    1.02
     information
    0.97
     time
    0.91
    emplar
    0.87
    ⅘
    0.85
     상세
    0.84
    just
    0.82
     Besøkt
    0.82
     itself
    0.80
     preparing
    0.79
    Activations Density 0.005%

    No Known Activations