© 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. 5096
    Prev
    Next
    INDEX
    Explanations

    urgent matters needed

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

    The neuron fires most strongly on formal, domain-specific noun phrases—like program or project titles and technical terms in environmental, health, or organizational contexts.

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

    technical or formal language related to programs, services, policies, or organizational initiatives addressing social needs or problems.

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

    formal, institutional or technical terminology—abstract nouns naming programs, services, resources, conditions, technologies, or specifications in bureaucratic/scientific contexts.

    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
    あるいは
    0.82
    
    0.66
    したり
    0.64
    或是
    0.63
    ก็
    0.61
    することも
    0.61
    หรือ
    0.61
    或
    0.60
    占い
    0.59
     或
    0.59
    POSITIVE LOGITS
     Category
    0.81
     ubicada
    0.77
     tại
    0.74
     Kabupaten
    0.73
     avec
    0.71
     महिला
    0.70
     yaitu
    0.70
     concernant
    0.70
    令和
    0.70
     Sonoma
    0.70
    Activations Density 0.021%

    No Known Activations