© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 16121
    Prev
    Next
    INDEX
    Explanations

    concepts in multiple languages

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

    The neuron spikes on the very last token of each short document snippet—it’s basically a “end‐of‐segment” detector.

    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
    る
    0.81
     PBMCs
    0.80
     специальных
    0.80
     glutathione
    0.77
     المعادلات
    0.77
     Meks
    0.77
     Transvaal
    0.76
    awara
    0.75
     lymphocyte
    0.75
    𝚋
    0.75
    POSITIVE LOGITS
    edged
    0.82
     ambito
    0.77
    ти
    0.76
    setColor
    0.76
     чле
    0.73
     echar
    0.72
     COUN
    0.71
     qualific
    0.71
    ливо
    0.70
     edged
    0.70
    Activations Density 0.000%

    No Known Activations