© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 71037
    Prev
    Next
    INDEX
    Explanations

    conjunctions and introductory phrases

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

    The neuron detects sentence‐initial discourse markers or transition words (e.g. “As,” “But,” “So,” “Instead,” “Now,” etc.).

    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
     and
    -2.19
     as
    -1.53
    routeProvider
    -1.48
     at
    -1.47
     during
    -1.42
     in
    -1.40
     pequeño
    -1.36
    نيف
    -1.34
    >--}}
    -1.33
    yyb
    -1.32
    POSITIVE LOGITS
     
    1.39
    5
    1.27
    %
    1.24
    0
    1.23
     Either
    1.20
    シャワー
    1.20
    如同
    1.16
     chim
    1.16
     verhind
    1.14
    一时间
    1.14
    Activations Density 0.163%

    No Known Activations