© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 2948
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on substantive content words—especially technical measures, qualifiers, and descriptive terms (e.g. frequencies, directions, adverbs like “smoothly” or “especially,” adjectives like “additional,” and domain‐specific nouns)—i.e. it’s keyed to informative, non‐function words in expository/technical 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
    am
    0.81
    _
    0.75
    UFF
    0.73
     VIDA
    0.70
    ats
    0.69
    ud
    0.68
     Photos
    0.67
     przej
    0.67
    aq
    0.66
     CIV
    0.66
    POSITIVE LOGITS
    Như
    0.83
    Ẫ
    0.79
    ందరు
    0.78
     gerais
    0.77
     договор
    0.77
     verwend
    0.77
    ర్టు
    0.77
    Anno
    0.77
    Además
    0.75
    Điều
    0.75
    Activations Density 0.000%

    No Known Activations