© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 98170
    Prev
    Next
    INDEX
    Explanations

    India can diplomatically

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

    The neuron lights up on named-entity tokens—proper names (people, organizations, places) and numeric/date tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    ׇ
    -1.32
     étan
    -1.27
     chré
    -1.21
    Exemples
    -1.15
    Voir
    -1.13
    루
    -1.10
    onView
    -1.09
    ۟
    -1.09
     also
    -1.08
     chrétienne
    -1.08
    POSITIVE LOGITS
     it
    1.39
     zrobić
    1.16
     美女
    1.09
     музыки
    1.09
    ּוֹ
    1.02
     sấy
    1.02
    itively
    1.00
    广场
    1.00
    ִּ
    0.98
     uscita
    0.97
    Activations Density 0.077%

    No Known Activations