© 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. 22-GEMMASCOPE-RES-131K
    7. 128810
    Prev
    Next
    INDEX
    Explanations

    This neuron activates on tokens that are parts of place names (e.g. city, town, or region names).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    punctuation and foreign words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    の基本
    -0.96
    -&
    -0.96
    vanou
    -0.96
     nevera
    -0.95
     Tercera
    -0.91
     tarta
    -0.91
     Comunidad
    -0.91
    थ
    -0.89
     mariposa
    -0.88
     permitan
    -0.88
    POSITIVE LOGITS
    二维码
    0.94
     beslu
    0.93
    汙
    0.91
     тис
    0.90
     грамм
    0.90
    อื่น
    0.90
    禮物
    0.89
     innumerable
    0.89
     النف
    0.88
    használ
    0.88
    Activations Density 0.031%

    No Known Activations