© 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. 34-GEMMASCOPE-RES-131K
    7. 92999
    Prev
    Next
    INDEX
    Explanations

    categories and citations

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

    The neuron strongly activates on standalone numeric strings and codes (e.g. years, page‐ or issue numbers, IDs), essentially flagging tokens composed of digits.

    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
    </h6>
    -0.91
    todoList
    -0.83
    -->
    -0.79
    toDo
    -0.75
    </h4>
    -0.74
    hov
    -0.73
     għall
    -0.73
     Rolle
    -0.72
    erdings
    -0.72
    ><!--
    -0.72
    POSITIVE LOGITS
     bushel
    0.82
    묻
    0.74
     largeur
    0.72
    ksis
    0.71
     transcendental
    0.71
    cutaneous
    0.70
    からは
    0.70
     Armies
    0.70
     発送
    0.70
    蘖
    0.69
    Activations Density 0.000%

    No Known Activations