© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9662
    Prev
    Next
    INDEX
    Explanations

    The neuron responds strongly to broad category or domain labels—single words naming high-level classes such as industry, country, region, species, etc.

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

    industries and organizations

    np_max-act-logits · claude-4-5-sonnetTriggered by @jrosseruk
    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
     multinational
    1.00
    ic
    0.82
    们
    0.82
     Canadi
    0.80
     transnational
    0.79
     descendant
    0.75
    ж
    0.75
     fide
    0.74
     Canadian
    0.73
     billion
    0.71
    POSITIVE LOGITS
    Wide
    0.86
     اولى
    0.78
    বিস্ত
    0.77
    之力
    0.75
     کیلئے
    0.74
    entwicklung
    0.74
    ০
    0.74
    ἀ
    0.74
     Announces
    0.73
    用来
    0.73
    Activations Density 0.976%

    No Known Activations