© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 105520
    Prev
    Next
    INDEX
    Explanations

    documentclass} \usepackage

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

    the opening curly brace after `\documentclass` in LaTeX code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron strongly activates on the LaTeX command that loads the “mdframed” package with the TikZ frame method (i.e. the line `\usepackage[frame method=TikZ]{mdframed}`).

    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
    了过来
    -1.55
     Coloque
    -1.54
     rique
    -1.45
     raiſ
    -1.42
     sintomas
    -1.42
    ྄
    -1.41
    櫸
    -1.41
    蓠
    -1.38
     celles
    -1.38
    DockStyle
    -1.38
    POSITIVE LOGITS
     this
    1.61
     Because
    1.55
     on
    1.39
    ŚĆ
    1.38
     for
    1.36
    かというと
    1.34
     From
    1.32
     September
    1.30
     November
    1.29
     all
    1.23
    Activations Density 0.000%

    No Known Activations