© 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. 33818
    Prev
    Next
    INDEX
    Explanations

    copyright notices and code headers

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

    This neuron detects programming‐style code constructs—especially import/include directives and related code/header tokens.

    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
     perce
    -1.05
     delightful
    -1.05
     amongst
    -0.95
     stronę
    -0.94
     nonché
    -0.91
    abha
    -0.89
    というか
    -0.89
    domada
    -0.88
    ウザ
    -0.87
    ãos
    -0.85
    POSITIVE LOGITS
     että
    1.04
    PUTY
    0.99
    nameof
    0.95
    häng
    0.93
     WITH
    0.91
    ;
    0.91
    .
    0.90
    員の
    0.87
    ксана
    0.86
     były
    0.86
    Activations Density 0.002%

    No Known Activations