© 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. 6476
    Prev
    Next
    INDEX
    Explanations

    joining text segments

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

    The neuron identifies section or article titles and headings, especially those that begin a new segment or topic.

    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
    chrome
    0.76
    েলার
    0.73
    লার
    0.73
    которые
    0.66
    窿
    0.66
    ної
    0.65
    ktion
    0.65
    िएशन
    0.65
     cream
    0.64
    ния
    0.63
    POSITIVE LOGITS
     nCount
    0.79
     continueRoutine
    0.75
     עבור
    0.75
     ברי
    0.75
     Primarily
    0.73
     אני
    0.71
     tạp
    0.70
     vivent
    0.70
     மன
    0.70
     由
    0.70
    Activations Density 0.000%

    No Known Activations