© 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. 8361
    Prev
    Next
    INDEX
    Explanations

    shortcodes or html tags

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

    This neuron fires on non‐natural text “markup” tokens—things like emoticon codes (e.g. :roll_eyes:, :thumbsup:), bracketed tags or captions, underscores in IDs/links, and other metadata 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_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
     estavam
    -0.80
     veio
    -0.78
     fizeram
    -0.75
    intahan
    -0.74
     ficaram
    -0.72
     trzeba
    -0.71
    vised
    -0.71
    arrived
    -0.71
    علن
    -0.69
    runOnUiThread
    -0.69
    POSITIVE LOGITS
     reson
    0.81
    冕
    0.76
     sect
    0.73
    Lef
    0.73
     diab
    0.72
    لاف
    0.71
     conversions
    0.70
    brei
    0.70
    маю
    0.69
     mapper
    0.69
    Activations Density 0.175%

    No Known Activations