© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 100740
    Prev
    Next
    INDEX
    Explanations

    LaTeX commands and syntax

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

    opening curly braces in LaTeX commands.

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

    The neuron is specialized to detect numeric strings (digits) appearing as arguments inside LaTeX command braces.

    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
    -
    -2.63
    o
    -2.25
    ar
    -2.14
    f
    -2.14
    t
    -1.93
    et
    -1.88
    h
    -1.88
     gedacht
    -1.79
      
    -1.77
     was
    -1.76
    POSITIVE LOGITS
    媗
    2.23
    憮
    2.16
    
    2.11
    捽
    2.06
    lgari
    2.05
    %");
    2.02
    蠵
    2.00
    %")
    1.87
    初心
    1.86
     Jér
    1.86
    Activations Density 0.017%

    No Known Activations