© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 116421
    Prev
    Next
    INDEX
    Explanations

    language and programming languages

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

    The neuron is triggered by LaTeX‐style math markup (dollar signs, mathcal, math symbols, etc.) indicating a mathematical expression.

    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
     Latency
    -0.86
    latency
    -0.72
     Filipina
    -0.71
    pruchs
    -0.69
     Clam
    -0.69
    vada
    -0.68
     الشب
    -0.68
    jmu
    -0.67
    coes
    -0.66
     Bavaria
    -0.63
    POSITIVE LOGITS
     language
    6.09
     languages
    4.84
    language
    4.22
    Language
    3.75
     Language
    3.75
    语言
    3.64
     Languages
    3.44
     LANGUAGE
    3.36
    languages
    3.28
    LANGUAGE
    3.22
    Activations Density 0.150%

    No Known Activations