© 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. 8660
    Prev
    Next
    INDEX
    Explanations

    improvements and

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron activates primarily on named-entity tokens—such as person names, place names, organizations, dates, and other proper nouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    proper nouns, especially place names and geographic locations.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    This neuron detects capitalized proper nouns and named entities (people, places, organizations/brands, dates, model names, acronyms, and other specific titles).

    oai_token-act-pair · gpt-5Triggered 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
     a
    0.63
    0.62
     also
    0.58
     the
    0.57
     an
    0.54
     =
    0.49
     ook
    0.49
    also
    0.49
     was
    0.49
     one
    0.47
    POSITIVE LOGITS
    istischen
    0.50
    rischen
    0.50
    cellaneous
    0.47
    <unused89>
    0.46
    ാത്ത
    0.46
    ukary
    0.46
    forschung
    0.44
    oglycos
    0.44
    ukiy
    0.44
    łat
    0.44
    Activations Density 0.916%

    No Known Activations