© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 12446
    Prev
    Next
    INDEX
    Explanations

    specific identities and roles

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

    The neuron selectively fires on mentions of prominent named entities—people, professional roles, institutions, and places.

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

    professional titles and proper nouns, particularly names of people, places, and organizations.

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

    prominent content nouns referring to institutions, professions, places, and technical or concrete objects (news-style keywords).

    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
     (
    0.71
     muß
    0.69
     darn
    0.66
     ALL
    0.63
     التى
    0.62
    (
    0.62
    ALL
    0.59
      
    0.59
     pudiera
    0.59
    President
    0.57
    POSITIVE LOGITS
     differed
    0.85
     onboarding
    0.80
     impactful
    0.75
     yaşad
    0.74
     downsides
    0.74
     farklı
    0.71
     downside
    0.71
     ifad
    0.70
     relatable
    0.69
     differs
    0.69
    Activations Density 0.010%

    No Known Activations