© 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. 20337
    Prev
    Next
    INDEX
    Explanations

    road and map, safety, or block

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

    The neuron fires on any token containing the substring “road” (e.g. “Road,” “road,” or “Roadmap”).

    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
    くちゃ
    -3.11
    vantaged
    -2.77
    就像是
    -2.72
    ﾏ
    -2.64
    也算是
    -2.61
    ℟
    -2.47
     また
    -2.39
    ｿ
    -2.33
    setBold
    -2.30
    ｺ
    -2.27
    POSITIVE LOGITS
     '
    3.06
     subtleties
    2.75
    an
    2.73
     mishap
    2.67
     Its
    2.52
     wither
    2.50
    ſam
    2.39
     emeritus
    2.39
     unsurpassed
    2.38
     definitive
    2.30
    Activations Density 0.012%

    No Known Activations