© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 13726
    Prev
    Next
    INDEX
    Explanations

    markup and code syntax

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

    The neuron primarily activates on LaTeX command tokens (i.e. backslash-initiated commands and package or environment declarations).

    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
    -
    -1.91
     stärken
    -1.82
    ‌.
    -1.75
     aliados
    -1.75
     êtes
    -1.73
     pueden
    -1.73
    潆
    -1.64
     tärke
    -1.59
     įva
    -1.59
     then
    -1.59
    POSITIVE LOGITS
     and
    2.58
    ↵↵
    2.14
     
    1.45
    malıdır
    1.45
    ější
    1.44
    の為
    1.41
    micamente
    1.40
    ＄
    1.38
    玩具
    1.38
     regalar
    1.37
    Activations Density 0.002%

    No Known Activations