© 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. 3273
    Prev
    Next
    INDEX
    Explanations

    former/also/nicknamed

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

    The neuron selectively responds to capitalized tokens that are part of proper names or titled entities (e.g., person names or work titles).

    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
    ──
    0.69
    すなわち
    0.68
    そして
    0.67
    namely
    0.67
    ，
    0.65
    なのですが
    0.64
    ち
    0.61
     cioè
    0.61
    っ
    0.60
    において
    0.59
    POSITIVE LOGITS
     ALSO
    0.81
     exudes
    0.80
     també
    0.71
     يمكنك
    0.70
     also
    0.70
     STARTED
    0.69
     sürekli
    0.68
     priced
    0.67
     길이가
    0.67
     مدت
    0.66
    Activations Density 0.073%

    No Known Activations