© 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. 1046
    Prev
    Next
    INDEX
    Explanations

    basic, limited, or traditional qualifiers

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

    The neuron is essentially a “segment-boundary” detector—it spikes whenever a new document or review chunk begins, marking the transition into a fresh text block.

    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
     аксессу
    0.88
     sekal
    0.88
     ویژگی
    0.82
    ﻼ
    0.82
     врач
    0.80
     erinner
    0.80
     همچ
    0.77
     выступа
    0.77
     появляется
    0.77
     buen
    0.76
    POSITIVE LOGITS
    work
    0.84
    server
    0.83
    lessly
    0.82
    sw
    0.79
    tests
    0.77
    he
    0.77
     August
    0.77
    l
    0.77
    truncate
    0.76
    justice
    0.76
    Activations Density 0.001%

    No Known Activations