© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 75674
    Prev
    Next
    INDEX
    Explanations

    start of LaTeX environments

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

    The neuron spikes on the LaTeX “begin” keyword, i.e. it detects the start of environments (the “\begin” token).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    c
    -1.89
    r
    -1.85
     more
    -1.84
    se
    -1.67
    lo
    -1.67
     will
    -1.64
    </td>
    -1.64
    e
    -1.62
     if
    -1.60
    nością
    -1.58
    POSITIVE LOGITS
    <bos>
    1.94
     Encuentra
    1.78
    kệ
    1.72
     lámparas
    1.69
    dịch
    1.67
     入力
    1.67
     manualidades
    1.65
     ใหม่
    1.63
     Introducción
    1.62
    ☐
    1.62
    Activations Density 0.001%

    No Known Activations