© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 382
    Prev
    Next
    INDEX
    Explanations

    This neuron detects prominent thematic or topic words (e.g., key subjects and section‐heading nouns) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    0
    0.48
    '
    0.45
     strangle
    0.41
     thermal
    0.40
     clipping
    0.40
     cantilever
    0.39
     perforation
    0.39
     grid
    0.39
    ０
    0.39
     average
    0.38
    POSITIVE LOGITS
    ictionaries
    0.51
     budaya
    0.51
    ಞ
    0.51
    iquetas
    0.50
     as
    0.49
    iritual
    0.49
     sebagai
    0.49
     katika
    0.49
     Sebagai
    0.48
     ética
    0.48
    Activations Density 0.000%

    No Known Activations