© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65854
    Prev
    Next
    INDEX
    Explanations

    LaTeX commands and document structures

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

    The neuron detects mentions of LaTeX package names (e.g. amsmath, tikz, pgfplots) in usepackage commands.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     on
    -2.06
     from
    -1.95
    providedIn
    -1.83
     it
    -1.75
     any
    -1.74
     to
    -1.63
     at
    -1.63
     those
    -1.58
    内心
    -1.57
     for
    -1.53
    POSITIVE LOGITS
     rapporter
    1.62
    <bos>
    1.50
     diriger
    1.48
     Nha
    1.48
    Bli
    1.42
    焢
    1.41
     opérateur
    1.41
     faciliter
    1.39
     jaunes
    1.37
     grotte
    1.36
    Activations Density 0.000%

    No Known Activations