© 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. 22-GEMMASCOPE-RES-131K
    7. 71414
    Prev
    Next
    INDEX
    Explanations

    latex math environment

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

    The neuron is specialized to detect LaTeX/math-mode constructs—especially numeric tokens and math delimiters (like “$$”, “(”, “)”, “*”, “quad”) in mathematical expressions.

    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
    providedIn
    -1.29
     Tiempo
    -1.09
    ⑶
    -1.09
    |_{
    -1.05
    )/
    -1.05
     then
    -1.05
    洐
    -1.02
    Polecam
    -1.01
    deepcopy
    -0.99
    Dzień
    -0.98
    POSITIVE LOGITS
    Istorija
    1.16
     amounts
    1.08
     fantástica
    1.07
    [
    1.01
     今
    0.98
     crystallized
    0.98
     солдат
    0.95
     diciendo
    0.95
     sanitario
    0.95
     piccola
    0.94
    Activations Density 0.003%

    No Known Activations