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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 70350
    Prev
    Next
    INDEX
    Explanations

    code imports and modules

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

    The neuron fires on uppercase identifier tokens—particularly configuration‐style constant names (e.g. DB_HOST, DB_USER, DB_NAME) and other all‐caps or CamelCaps parameters.

    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
     prestigio
    -0.91
    rogly
    -0.83
     дописавши
    -0.83
     eventualmente
    -0.82
    кту
    -0.81
    itel
    -0.81
    ambito
    -0.81
    ブログ村
    -0.81
    Lng
    -0.77
     Drummond
    -0.77
    POSITIVE LOGITS
     their
    1.02
    SetBool
    0.94
     services
    0.91
    muerzo
    0.88
     biografias
    0.87
    ⅱ
    0.85
    車が
    0.85
    車は
    0.84
    soundcloud
    0.83
     modules
    0.82
    Activations Density 0.003%

    No Known Activations