© 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. 6921
    Prev
    Next
    INDEX
    Explanations

    file type specifications

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

    This neuron selectively activates on capitalized words or proper names (typically brand or series names and sentence-initial uppercase tokens).

    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
    u
    0.80
    雰囲気
    0.73
    eldo
    0.72
    បែប
    0.71
    ેલ
    0.70
    有多少
    0.70
    ópolis
    0.70
    cape
    0.69
    يروس
    0.69
    扱
    0.69
    POSITIVE LOGITS
     Feminist
    0.82
     размере
    0.80
    меры
    0.80
    inde
    0.80
    ה
    0.80
    пута
    0.79
    ран
    0.78
    ложить
    0.75
     দেবার
    0.75
    चलिए
    0.75
    Activations Density 0.001%

    No Known Activations