© 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. 116486
    Prev
    Next
    INDEX
    Explanations

    mathematical questions and calculations

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

    This neuron detects numeric content—including digit words (“digit”), numerals (e.g. “2,” “7,” “10,000”), place-value terms (“thousands,” “hundreds”), and arithmetic operators (–, *, etc.)—in headings, math problems, and titles.

    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
     interesado
    -0.77
     eius
    -0.77
    jangan
    -0.75
    illants
    -0.73
    vtx
    -0.71
    signia
    -0.70
    fractive
    -0.69
     kepem
    -0.68
    vorg
    -0.68
    Mongo
    -0.68
    POSITIVE LOGITS
     mental
    1.05
     facts
    1.00
     Mental
    0.87
    Computation
    0.87
     computation
    0.87
     mentally
    0.87
     fluency
    0.87
     Algorithm
    0.84
     Computation
    0.84
     fact
    0.84
    Activations Density 0.041%

    No Known Activations