© 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. 102861
    Prev
    Next
    INDEX
    Explanations

    welcome

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

    The neuron strongly activates on the kinds of words you find in promotional or call‐to‐action copy—words like “on,” “with,” “welcome,” “drive,” and similar—that link a headline or slogan to its pitch.

    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
    -1.75
     semua
    -1.73
     aDecoder
    -1.63
     மற்றும்
    -1.55
     HUGE
    -1.48
    ！"
    -1.43
     sudah
    -1.41
    」、「
    -1.41
     giving
    -1.41
     an
    -1.41
    POSITIVE LOGITS
    냑
    1.54
    がありますが
    1.53
     you
    1.48
    )
    1.47
    even
    1.43
    あなたが
    1.42
    particularly
    1.41
    揍
    1.38
     convierte
    1.37
     largely
    1.34
    Activations Density 0.091%

    No Known Activations