© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 71459
    Prev
    Next
    INDEX
    Explanations

    continuous improvement and evolution

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

    The neuron fires on words and forms that signal improvement or positive change (e.g. improve/​improving, enhanced, evolving, continuously, update).

    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
     некоторых
    -1.41
    ulite
    -1.38
     некоторы
    -1.32
    的一些
    -1.27
        
    -1.26
     mirando
    -1.23
     большинстве
    -1.22
     ciem
    -1.21
    gorithm
    -1.20
    够了
    -1.20
    POSITIVE LOGITS
     our
    1.59
    new
    1.41
    re
    1.30
     приготовить
    1.27
     totalité
    1.24
     CONCLUS
    1.23
     ========
    1.22
    NDICE
    1.20
     mientras
    1.19
     baju
    1.19
    Activations Density 0.031%

    No Known Activations