© 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. 1597
    Prev
    Next
    INDEX
    Explanations

    urgency, detection, upset, ill, fear

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

    The neuron strongly activates on tokens relating to legal or regulatory discussion (e.g. legality status, government bodies, restrictions).

    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
    XVII
    0.48
    ურთიერთ
    0.46
    ುತ್ತೇನೆ
    0.46
    が出来
    0.46
    xiii
    0.45
    ১৭
    0.44
    OECD
    0.44
    ೩
    0.44
    vii
    0.43
    cailles
    0.43
    POSITIVE LOGITS
     dentist
    0.42
     payday
    0.42
     malfunction
    0.41
     check
    0.41
     dental
    0.40
     malfunctions
    0.40
     among
    0.40
     Keto
    0.40
     Dentist
    0.40
     Your
    0.40
    Activations Density 0.054%

    No Known Activations