© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 76189
    Prev
    Next
    INDEX
    Explanations

    New York

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

    This neuron specifically detects the capitalized word “New,” as it appears in names like “New York Times,” “New Yorker,” etc.

    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
    Opis
    -0.80
     شهادت
    -0.74
    Zab
    -0.72
     improvisation
    -0.71
    ünstig
    -0.70
     Genevieve
    -0.69
     Νο
    -0.69
    Tú
    -0.69
     corpo
    -0.69
    hướng
    -0.69
    POSITIVE LOGITS
     York
    0.84
    iltà
    0.80
    cienti
    0.77
    Guard
    0.77
     Scientist
    0.76
    Review
    0.75
    ticals
    0.75
     Guard
    0.74
    oluted
    0.73
    Confirm
    0.73
    Activations Density 0.034%

    No Known Activations