© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 130684
    Prev
    Next
    INDEX
    Explanations

    The neuron is primarily detecting literal constants in code (numeric literals and Boolean constants).

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

    boolean and bool types in code

    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
    ophe
    -0.95
    婠
    -0.93
    Recommend
    -0.91
    DMETHOD
    -0.88
     także
    -0.88
     опыта
    -0.88
    richten
    -0.86
     også
    -0.86
     alemanes
    -0.86
     traduire
    -0.85
    POSITIVE LOGITS
     try
    0.96
    المصادر
    0.96
     lusso
    0.93
     tried
    0.92
     check
    0.89
    if
    0.86
     attempt
    0.85
    try
    0.84
     링
    0.83
    انظر
    0.82
    Activations Density 0.004%

    No Known Activations