© 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. 3379
    Prev
    Next
    INDEX
    Explanations

    multilingual context processing

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

    The neuron fires on words that occur right at the start of a new segment (i.e. the first content token after a <bos>), effectively flagging sentence-initial content words.

    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
     gesamte
    0.71
    entum
    0.70
    Segmentation
    0.68
     جائزے
    0.68
    temperatur
    0.68
    тивность
    0.66
    ᱶ
    0.65
    adati
    0.64
    ᆸ
    0.64
    iž
    0.64
    POSITIVE LOGITS
    国家
    0.80
     PO
    0.79
    אם
    0.79
    lar
    0.77
     risc
    0.75
    电视台
    0.75
    لي
    0.75
     Chicks
    0.74
    או
    0.74
     ora
    0.72
    Activations Density 0.000%

    No Known Activations