© 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. 67205
    Prev
    Next
    INDEX
    Explanations

    response

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

    The neuron is keyed to punctuation and numeric‐style tokens—things like commas, colons, dashes, lines of “=”, and standalone numbers (e.g. timestamps or separators).

    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
    स्था
    -0.83
    javax
    -0.83
    ocere
    -0.83
     Recom
    -0.82
    jutkan
    -0.81
    WebSocket
    -0.79
     currentDate
    -0.79
    otide
    -0.78
    iali
    -0.77
    kitan
    -0.77
    POSITIVE LOGITS
     response
    2.19
     responded
    2.09
     reply
    2.05
     respond
    1.83
     replied
    1.75
     replies
    1.70
     responses
    1.67
    RESPONSE
    1.66
    reply
    1.66
    response
    1.66
    Activations Density 0.074%

    No Known Activations