© 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. 24-GEMMASCOPE-2-RES-16K
    7. 10000
    Prev
    Next
    INDEX
    Explanations

    This neuron consistently fires on the very first substantive word of a new text block or 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
    正如
    0.89
     Tourists
    0.82
    感激
    0.82
    ों
    0.79
    防护
    0.77
    Мен
    0.77
    
    0.77
    Passengers
    0.77
     missiles
    0.73
    Chúc
    0.73
    POSITIVE LOGITS
    an
    0.90
    in
    0.88
    o
    0.86
    is
    0.83
    e
    0.83
    ein
    0.82
    a
    0.82
    oft
    0.81
    un
    0.80
    on
    0.79
    Activations Density 0.002%

    No Known Activations