© 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. 34-GEMMASCOPE-RES-131K
    7. 83386
    Prev
    Next
    INDEX
    Explanations

    code, html, status, data

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

    The neuron flags programming-language code syntax—e.g. function calls, braces/parentheses, keywords like “function” or jQuery “$(…)” lines.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     venerable
    -1.22
     jagged
    -1.20
    čnej
    -1.18
     laudable
    -1.13
    ඹ
    -1.13
     garota
    -1.09
     undetermined
    -1.09
     mingle
    -1.09
    liski
    -1.07
    Eksterne
    -1.05
    POSITIVE LOGITS
     dage
    1.04
     cepillo
    1.02
     брюки
    1.01
     foi
    1.01
     zand
    1.00
    ૮
    1.00
    𝜋
    1.00
     monst
    0.99
     lande
    0.98
     tuyo
    0.98
    Activations Density 0.037%

    No Known Activations