© 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. 35554
    Prev
    Next
    INDEX
    Explanations

    starts questions or code diffs

    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
     or
    -1.65
     There
    -1.59
    скажите
    -1.45
     reduce
    -1.42
     prevent
    -1.39
    をしました
    -1.37
    There
    -1.35
     Those
    -1.34
     like
    -1.34
     When
    -1.34
    POSITIVE LOGITS
     Dinas
    1.49
    があり
    1.48
     ß
    1.44
    躥
    1.41
    鶿
    1.31
     temperat
    1.30
     '*',
    1.30
     elec
    1.29
    푆
    1.28
     tyl
    1.27
    Activations Density 0.001%

    No Known Activations