© 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. 59829
    Prev
    Next
    INDEX
    Explanations

    variance and covariance

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

    The neuron specializes in spotting LaTeX‐style mathematical tokens—variables, operators, and function names—within formula notation.

    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
    ׇ
    -0.80
    jint
    -0.76
    iLocal
    -0.75
     musée
    -0.73
    useum
    -0.70
    Universal
    -0.70
    學院
    -0.70
    lutning
    -0.70
     iLocal
    -0.69
    httpd
    -0.69
    POSITIVE LOGITS
     variance
    4.44
     variances
    3.63
     Variance
    3.55
    variance
    3.39
    Variance
    3.30
     Var
    2.52
     var
    2.45
    Var
    2.27
     Vari
    2.25
     covariance
    2.25
    Activations Density 0.053%

    No Known Activations