© 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. 98526
    Prev
    Next
    INDEX
    Explanations

    mathematical and code symbols

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

    The neuron fires on programming/code tokens—especially numeric literals and identifier names in code snippets.

    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
    报记者
    -1.09
    上げます
    -1.01
    それにしても
    -1.00
    ？？？
    -0.99
    <bos>
    -0.98
    の中で
    -0.98
    などの
    -0.98
    买的
    -0.96
    游戏特色
    -0.96
     chaîne
    -0.94
    POSITIVE LOGITS
    ところに
    1.15
    感じが
    1.09
     And
    1.09
     What
    1.06
     his
    1.05
    なのかな
    1.01
    気がする
    1.00
    なもの
    1.00
    有三
    0.99
    Should
    0.98
    Activations Density 0.044%

    No Known Activations