© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 77257
    Prev
    Next
    INDEX
    Explanations

    specific word completions

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

    The neuron strongly activates on uppercase acronyms, punctuation (slashes, asterisks, periods), and other formatting tokens typical of code comments or license‐header boilerplate.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     brigade
    -0.75
    umbria
    -0.74
    ➡️
    -0.74
    PARTICIP
    -0.73
    لیت
    -0.73
    вів
    -0.73
    gand
    -0.72
     Jamb
    -0.72
     экспер
    -0.71
    Hyundai
    -0.71
    POSITIVE LOGITS
     citizen
    0.82
     Industries
    0.75
     buildings
    0.74
     вод
    0.73
     zoned
    0.71
    渭
    0.70
     demand
    0.70
     peeps
    0.69
     CS
    0.69
     query
    0.69
    Activations Density 0.020%

    No Known Activations