© 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. 35634
    Prev
    Next
    INDEX
    Explanations

    mathematical formulas and integrals

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

    This neuron identifies LaTeX math operator and function tokens (e.g. sum, log, frac, subscripts) within formula sections.

    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
     any
    -1.39
     to
    -1.38
    。
    -1.38
     its
    -1.37
     also
    -1.33
     "
    -1.19
    （
    -1.16
    한
    -1.14
    ="
    -1.09
     و
    -1.09
    POSITIVE LOGITS
     parlar
    1.33
    Desta
    1.29
    BENEF
    1.24
     enthou
    1.22
     commer
    1.22
    Escrito
    1.21
    Seni
    1.20
    addPreferred
    1.20
    Nummer
    1.18
     poë
    1.18
    Activations Density 0.032%

    No Known Activations