© 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. 5398
    Prev
    Next
    INDEX
    Explanations

    The neuron is looking for the very first substantive (non-function) word at the start of a new paragraph.

    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
    ம்
    0.94
     coughing
    0.91
    ленной
    0.85
    лән
    0.81
    phoneNumber
    0.81
     tetragonal
    0.81
     Epistle
    0.79
     낙
    0.79
    ียน
    0.78
    తుంది
    0.78
    POSITIVE LOGITS
    $
    0.96
    NEW
    0.82
    Produced
    0.80
    Nature
    0.75
    Source
    0.73
    ////////////////
    0.73
    PORT
    0.73
    December
    0.70
    Sources
    0.69
    mojo
    0.68
    Activations Density 0.000%

    No Known Activations