© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 65019
    Prev
    Next
    INDEX
    Explanations

    Thai names and questions

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

    This neuron fires on roman-alphabet tokens that are parts of Thai names or Thai proper nouns.

    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
     leprosy
    -0.73
    排列
    -0.72
     Kurse
    -0.71
     colorful
    -0.71
    小龙
    -0.70
    Belt
    -0.69
     perilaku
    -0.69
    rumah
    -0.69
    haji
    -0.69
     neighboring
    -0.68
    POSITIVE LOGITS
    ching
    1.00
    킁
    0.97
    orn
    0.94
     Manor
    0.91
    keeps
    0.90
     întâ
    0.89
    簓
    0.89
     gallo
    0.89
    tifying
    0.88
    $_,
    0.88
    Activations Density 0.009%

    No Known Activations