© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67520
    Prev
    Next
    INDEX
    Explanations

    question and answer format

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

    The neuron strongly fires on second-person imperative or advisory language—phrases that directly address “you” and urge or guide the reader to take some action (e.g. “If you prefer to purchase an essay … you may want to think about …”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     červená
    -1.40
     when
    -1.38
     three
    -1.31
     modrá
    -1.31
     černá
    -1.28
     after
    -1.26
    xspace
    -1.23
     before
    -1.22
    baju
    -1.17
    kenzo
    -1.15
    POSITIVE LOGITS
    _
    1.34
     desenvolvimento
    1.32
     ridiculously
    1.27
     obstru
    1.21
     acestei
    1.20
     этом
    1.15
    $
    1.15
    is
    1.15
     noastre
    1.14
     multicolore
    1.13
    Activations Density 0.013%

    No Known Activations