© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 9656
    Prev
    Next
    INDEX
    Explanations

    This neuron detects the very first word of a new segment or sentence (the sentence-initial token).

    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
    да
    0.75
     Д
    0.71
     dhe
    0.71
     zemlje
    0.64
    ма
    0.63
    ב
    0.63
     crede
    0.62
     quarks
    0.62
     О
    0.61
     पदों
    0.61
    POSITIVE LOGITS
    伲
    0.75
    衔
    0.73
    یداری
    0.72
    ม่า
    0.70
    投诉
    0.69
    现
    0.69
    回头
    0.69
    长
    0.69
    普遍
    0.68
    نيه
    0.68
    Activations Density 0.009%

    No Known Activations