© 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. 12494
    Prev
    Next
    INDEX
    Explanations

    verb at end of sentence

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

    The neuron fires on formal regulatory and legal terminology around insurance (e.g. “life insurance,” “regulating body,” “obligations,” “branch,” “corporations,” “law”), i.e. it looks for insurance-regulation language.

    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
     the
    0.64
     bề
    0.60
     Roma
    0.59
     equatorial
    0.58
     XRD
    0.57
     fortifications
    0.57
     tubercul
    0.56
    英
    0.55
     morphological
    0.55
     mitra
    0.55
    POSITIVE LOGITS
    ของ
    0.93
     που
    0.86
    ของการ
    0.85
     ভীষণ
    0.82
     ของ
    0.80
     של
    0.79
     که
    0.79
     của
    0.79
     التي
    0.78
    របស់
    0.77
    Activations Density 0.015%

    No Known Activations