© 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. 78332
    Prev
    Next
    INDEX
    Explanations

    proper nouns and foreign words

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

    The neuron sharply activates on high-level section or document headings and titles—e.g. HTML/XML <head> or <title> tags, film titles, organization names, and other prominent proper-noun labels.

    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
    ristas
    -0.99
    รรม
    -0.90
    animaux
    -0.88
    τικός
    -0.83
    šku
    -0.83
    Nand
    -0.80
    visualization
    -0.79
     DUR
    -0.77
     blessé
    -0.77
    paralle
    -0.77
    POSITIVE LOGITS
     activity
    0.96
    ก
    0.88
     formed
    0.86
     trở
    0.86
     đội
    0.81
    onset
    0.78
    append
    0.77
     founded
    0.77
     Hilfe
    0.76
     Drž
    0.75
    Activations Density 0.003%

    No Known Activations