© 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. 48736
    Prev
    Next
    INDEX
    Explanations

    classification of settlements and populations

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

    The neuron is primarily activating on numeric tokens—especially statistical or demographic figures (e.g. population counts, census years, percentages).

    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
     comparatively
    -1.06
     relatively
    -0.98
     efficiently
    -0.87
     paradigma
    -0.86
     surprisingly
    -0.86
     typically
    -0.85
     frequently
    -0.84
     deeply
    -0.82
     tami
    -0.82
     hvid
    -0.82
    POSITIVE LOGITS
     city
    0.88
     quintessential
    0.87
     designations
    0.87
     ян
    0.84
    wany
    0.83
     dingen
    0.83
    duate
    0.81
     goede
    0.81
     woes
    0.81
     cặp
    0.80
    Activations Density 0.005%

    No Known Activations