© 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. 4476
    Prev
    Next
    INDEX
    Explanations

    multilingual verbs/actions

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

    This neuron detects the Korean legal‐action verb “제기” (as in “제기하다,” meaning to file or bring a lawsuit).

    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
    を
    0.83
    に
    0.78
    や
    0.71
    の
    0.68
    です
    0.64
    を見
    0.64
    ズ
    0.62
     کے
    0.62
    等を
    0.61
    中
    0.61
    POSITIVE LOGITS
     QUAL
    0.70
     MINUTES
    0.66
     мүмк
    0.64
     여
    0.64
     REQUIREMENTS
    0.63
     삼성
    0.61
     Abstracts
    0.61
    했
    0.61
     Leute
    0.60
     Alters
    0.60
    Activations Density 0.000%

    No Known Activations