© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 12671
    Prev
    Next
    INDEX
    Explanations

    definitions and contexts

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

    This neuron fires on the very first content word of a new paragraph or section (i.e. the first token after a blank line or major break).

    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
    not
    -1.92
    at
    -1.82
    AN
    -1.78
    zembro
    -1.75
    "
    -1.73
    ‴
    -1.69
    which
    -1.67
    s
    -1.62
    ſelf
    -1.61
    with
    -1.60
    POSITIVE LOGITS
    櫪
    1.72
    rhosis
    1.64
    蚺
    1.56
     erreur
    1.54
    HexString
    1.52
    gucig
    1.52
    nisses
    1.52
     включа
    1.50
     prudent
    1.48
    神奇
    1.46
    Activations Density 0.000%

    No Known Activations