© 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. 34-GEMMASCOPE-RES-131K
    7. 34710
    Prev
    Next
    INDEX
    Explanations

    American Biographical Institute

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

    The neuron fires primarily on isolated technical identifiers—long alphanumeric tokens such as class or package names, version numbers, and similar code‐style or document‐reference IDs.

    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
     its
    -0.79
     vary
    -0.79
    掏
    -0.77
     careful
    -0.76
     wider
    -0.74
     certain
    -0.74
     giant
    -0.72
    edad
    -0.71
     forty
    -0.71
     larger
    -0.70
    POSITIVE LOGITS
     Bernd
    0.83
     Workmen
    0.81
    お世話
    0.81
    mbols
    0.79
     Adidas
    0.78
    єр
    0.78
     Innes
    0.78
    喹
    0.77
    ంక
    0.77
     puffed
    0.77
    Activations Density 0.000%

    No Known Activations