© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 44099
    Prev
    Next
    INDEX
    Explanations

    comes followed by context

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

    The neuron fires on the news‐style connective “comes” (as in “comes as” or “comes after”) used to introduce contextual developments.

    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
    😓
    -1.80
    🤣🤣🤣
    -1.76
    😡
    -1.73
    ǻ
    -1.70
    😖
    -1.70
    😫
    -1.68
     ),
    -1.67
    🤣🤣
    -1.67
    ☹
    -1.66
     ſmall
    -1.64
    POSITIVE LOGITS
     in
    1.95
     at
    1.88
    1
    1.80
    5
    1.71
     or
    1.63
     despite
    1.63
     after
    1.59
     did
    1.58
     during
    1.55
     before
    1.52
    Activations Density 0.019%

    No Known Activations