© 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. 103006
    Prev
    Next
    INDEX
    Explanations

    calculus and differential equations

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

    The neuron is triggered by chunks of mathematical notation—especially capital‐letter function names and their parentheses (e.g. “L(M)”) and related formulaic 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_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
     upgrade
    -0.71
    furter
    -0.71
     fija
    -0.69
     سبيل
    -0.67
     game
    -0.66
     fusion
    -0.65
    Fusion
    -0.65
    ubo
    -0.65
     permettra
    -0.65
     straw
    -0.64
    POSITIVE LOGITS
     Laplace
    1.57
     convolution
    1.21
    convolution
    1.00
    Lap
    0.97
     Convolution
    0.93
    Convolution
    0.90
    menubar
    0.84
    ℒ
    0.84
     Hea
    0.83
     оригинала
    0.81
    Activations Density 0.024%

    No Known Activations