© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1420
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on content-heavy, multisyllabic “information” words (longer, less frequent nouns and verbs) rather than on common short function words.

    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_31_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
    trong
    0.74
     ድረስ
    0.74
     עבור
    0.72
    問い合わせ
    0.67
    வராக
    0.61
    났
    0.61
    elerinde
    0.61
    asser
    0.60
    상은
    0.60
    ೂರು
    0.58
    POSITIVE LOGITS
     kepada
    3.42
     to
    3.36
     إلى
    3.02
     الى
    2.83
     thereto
    2.53
     إليه
    2.44
     إليها
    2.39
    到
    2.36
     aos
    2.13
    to
    2.03
    Activations Density 2.714%

    No Known Activations