© 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. 22-GEMMASCOPE-RES-131K
    7. 93125
    Prev
    Next
    INDEX
    Explanations

    names following 'and' or comma separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     Kombin
    -1.10
     deklar
    -1.01
     kompon
    -0.98
     Kompon
    -0.96
     Strukt
    -0.93
    Popis
    -0.90
    aktifkan
    -0.88
    üttel
    -0.87
    нева
    -0.86
     CONST
    -0.86
    POSITIVE LOGITS
    同样的
    1.16
    たびに
    1.11
    正常的
    1.09
     フー
    1.09
    普通的
    1.05
     afect
    1.05
     подрост
    1.05
     tremend
    1.05
     crea
    1.02
    如果不是
    1.02
    Activations Density 0.002%

    No Known Activations