© 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. 10-GEMMASCOPE-RES-131K
    7. 47750
    Prev
    Next
    INDEX
    Explanations

    base64 encoding and related concepts

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

    The main thing this neuron does is detect references to Base64 encoding (e.g. the “base”, “6”, “4”, and “encode” 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_10/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
    .
    -1.97
     For
    -1.84
    ↵↵
    -1.74
     onPostExecute
    -1.70
    </h5>
    -1.69
    BrowserModule
    -1.67
     generaciones
    -1.66
     what
    -1.66
     veinte
    -1.65
     have
    -1.63
    POSITIVE LOGITS
    with
    2.05
    秶
    1.78
    这事
    1.78
     technisch
    1.77
    未来的
    1.77
    biamo
    1.71
    你看
    1.70
    スーツ
    1.66
     Vám
    1.66
    ‌ها
    1.65
    Activations Density 0.003%

    No Known Activations