© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 71957
    Prev
    Next
    INDEX
    Explanations

    technical or mathematical context

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

    This neuron detects LaTeX commands and math‐environment syntax (backslash commands, environment delimiters, etc.).

    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
     dará
    -1.90
    などに
    -1.89
     seguirá
    -1.88
     hochwertige
    -1.82
     anuncia
    -1.79
     continuará
    -1.79
     alcuni
    -1.69
    が存在
    -1.64
     contará
    -1.63
     mismos
    -1.61
    POSITIVE LOGITS
    <em>
    1.87
    after
    1.84
     our
    1.83
    ?
    1.80
    i
    1.73
    Three
    1.70
    *
    1.67
    <b>
    1.65
     мою
    1.65
    as
    1.63
    Activations Density 0.000%

    No Known Activations