© 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. 15461
    Prev
    Next
    INDEX
    Explanations

    alias mapping, algorithm campaign, org packages

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

    The neuron fires on C/C++ header file tokens (the “.h” file names in include statements).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     numele
    -0.93
    Étymologie
    -0.90
    zeppelin
    -0.89
     borboleta
    -0.88
     sostu
    -0.85
    Perú
    -0.85
    ylen
    -0.85
    িতে
    -0.84
     približ
    -0.84
     šte
    -0.84
    POSITIVE LOGITS
     ymm
    0.96
    雾
    0.95
     sampai
    0.95
     should
    0.94
    Ра
    0.93
    ब
    0.93
    ロ
    0.90
    २
    0.88
     operations
    0.88
     circled
    0.88
    Activations Density 0.070%

    No Known Activations