© 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. 60196
    Prev
    Next
    INDEX
    Explanations

    return null or conditional logic

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

    This neuron predominantly activates on code punctuation and operator symbols (e.g. question marks, colons, angle brackets) rather than on alphanumeric text.

    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
     checkered
    -0.91
    佼
    -0.83
    出席
    -0.79
    多多
    -0.79
    atte
    -0.79
     percol
    -0.77
     einiger
    -0.74
    upakan
    -0.73
    ../../../
    -0.73
     unforgettable
    -0.73
    POSITIVE LOGITS
     <>
    1.16
     escuchadas
    0.94
     [
    0.84
    :]
    0.84
    <>
    0.81
     relojes
    0.79
    uisine
    0.79
    Fragment
    0.79
    <h4>
    0.78
     още
    0.78
    Activations Density 0.002%

    No Known Activations