© 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. 2462
    Prev
    Next
    INDEX
    Explanations

    action performed or item provided

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

    The neuron detects boilerplate disclaimer or attribution phrases indicating that information or materials are “provided,” “contained,” “featured,” or otherwise formally presented.

    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
     кто
    0.59
     чтобы
    0.59
     чтоб
    0.55
     για
    0.54
     когда
    0.52
     ナ
    0.52
    ر
    0.52
    ت
    0.52
     اپ
    0.52
     dla
    0.51
    POSITIVE LOGITS
     becomes
    0.74
     will
    0.70
     comprises
    0.68
     adalah
    0.67
     strives
    0.66
     consists
    0.66
     converges
    0.65
     achieves
    0.63
     differs
    0.62
     undergoes
    0.62
    Activations Density 0.002%

    No Known Activations