© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 7777
    Prev
    Next
    INDEX
    Explanations

    refusal due to ethical conflicts

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

    The neuron activates on interrogative sentences—i.e. it detects question words or question‐like phrasing.

    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_31_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
     весьма
    0.80
    这样一个
    0.74
    Cette
    0.73
     상당히
    0.68
    <b>
    0.67
    无论是
    0.67
    的那
    0.66
    <h4>
    0.66
    This
    0.66
    :)
    0.66
    POSITIVE LOGITS
     vaš
    0.76
     necesitas
    0.74
     schützen
    0.72
     YOUR
    0.71
     minha
    0.69
     kuhusu
    0.68
     puedes
    0.66
     ваша
    0.66
     goodbye
    0.66
     др
    0.66
    Activations Density 0.354%

    No Known Activations