© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 87211
    Prev
    Next
    INDEX
    Explanations

    image captions or file names

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

    The neuron strongly responds to LaTeX markup commands (tokens beginning with “\” and their associated keywords such as documentclass, usepackage, includegraphics, footnote, 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_34/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
    documentclass
    -0.88
    cdata
    -0.75
    &=
    -0.75
     document
    -0.75
     jedem
    -0.74
    cery
    -0.73
    Тогда
    -0.73
     Geden
    -0.72
     intelig
    -0.72
     menyebut
    -0.71
    POSITIVE LOGITS
     caption
    1.45
    caption
    1.41
     captions
    1.31
     Captions
    1.12
    Caption
    1.02
     Caption
    1.02
     captioned
    0.98
    figcaption
    0.84
    klär
    0.81
    föl
    0.81
    Activations Density 0.003%

    No Known Activations