© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 106241
    Prev
    Next
    INDEX
    Explanations

    Guinea, Bissau, Cayenne

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

    geographical locations and place names.

    oai_token-act-pair · claude-3-7-sonnet-20250219
    Triggered by @neilrathi

    The neuron fires strongly on capitalized named entities—especially geographic place names (e.g., countries, regions, parishes, parks).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    stop
    -0.71
    ahuila
    -0.71
     Cess
    -0.70
    tır
    -0.70
     Steff
    -0.69
    ᷅
    -0.69
    Ꭼ
    -0.69
     Decentral
    -0.69
     inquiries
    -0.68
     Drift
    -0.68
    POSITIVE LOGITS
     Guine
    1.03
    agré
    0.83
    ortion
    0.77
    anov
    0.76
     billeder
    0.76
     solicitation
    0.73
     Goldman
    0.73
    chter
    0.72
     آیفون
    0.72
     preservation
    0.72
    Activations Density 0.012%

    No Known Activations