© 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. 16355
    Prev
    Next
    INDEX
    Explanations

    language, state, technique, application

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

    The neuron responds strongly to mentions of named entities and key domain nouns—such as companies, products, countries, and other proper‐name terms.

    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
     c
    0.69
     US
    0.68
     I
    0.60
    ti
    0.58
    ↵↵
    0.58
     or
    0.57
     NO
    0.57
     Walt
    0.57
     FEC
    0.56
    ↵↵↵
    0.56
    POSITIVE LOGITS
     mantenerse
    0.73
     brainchild
    0.70
     jeho
    0.65
    ísimo
    0.65
     popolare
    0.64
     haría
    0.63
    を用いた
    0.63
     mencionado
    0.62
    唠
    0.62
    ющее
    0.61
    Activations Density 0.400%

    No Known Activations