© 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. 62780
    Prev
    Next
    INDEX
    Explanations

    addresses and locations

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

    The neuron fires on document segments that look like structured address fields—e.g. “Address: 1224” or “Location: HK Island: Central”—essentially spotting metadata labels and associated numeric or place-name address data.

    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
     család
    -0.77
     famiglia
    -0.75
     Ragnar
    -0.75
     baht
    -0.75
    sequences
    -0.74
     {}\
    -0.73
     orgull
    -0.73
    tops
    -0.73
    mouseup
    -0.72
    Orsay
    -0.72
    POSITIVE LOGITS
     Plot
    1.89
     plot
    1.79
    Plot
    1.67
    plot
    1.63
     Plots
    1.45
     plots
    1.40
     \#
    1.28
     address
    1.27
     house
    1.26
    PLOT
    1.26
    Activations Density 0.015%

    No Known Activations