© 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. 10-GEMMASCOPE-RES-131K
    7. 130164
    Prev
    Next
    INDEX
    Explanations

    financial and capital markets

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

    The neuron is primarily detecting non-Latin (especially Chinese) characters or text segments.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     religieuses
    -1.90
     chré
    -1.83
    ֩
    -1.80
     catholique
    -1.79
     katholischen
    -1.75
     この
    -1.68
     כוס
    -1.65
     этой
    -1.61
     ὁ
    -1.59
     włosów
    -1.56
    POSITIVE LOGITS
    ּוֹ
    3.03
    ִּ
    2.17
    ֶּ
    2.13
    וּ
    2.05
    ַּ
    2.05
     épaules
    1.98
    seekers
    1.91
    ogues
    1.89
    amino
    1.88
    ְּ
    1.73
    Activations Density 0.061%

    No Known Activations