© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 113967
    Prev
    Next
    INDEX
    Explanations

    the consequence

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

    This neuron responds strongly to the discourse marker “in turn,” flagging that specific phrase.

    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
    甞
    -2.17
    ViewById
    -1.81
    
    -1.75
    ’)
    -1.74
    ystemen
    -1.73
    畱
    -1.70
    prakti
    -1.67
    libatkan
    -1.66
    atı
    -1.63
    atiker
    -1.63
    POSITIVE LOGITS
    k
    2.39
    The
    2.28
    o
    2.20
    (
    2.13
    b
    2.05
    is
    2.03
    ll
    1.97
    ?
    1.93
    u
    1.91
    ed
    1.90
    Activations Density 0.008%

    No Known Activations