© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 69017
    Prev
    Next
    INDEX
    Explanations

    programming snippets

    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
     should
    -1.17
     already
    -1.17
    迅速
    -1.16
    서는
    -1.14
     is
    -1.05
    nees
    -1.02
     other
    -0.99
     certainly
    -0.97
    (;;)
    -0.96
    ᶦ
    -0.95
    POSITIVE LOGITS
    久し
    1.31
    licopter
    1.25
    すみません
    1.23
    dding
    1.20
    iress
    1.16
     graciosos
    1.11
    ähkö
    1.10
    ごめん
    1.09
    ebly
    1.09
     nélk
    1.09
    Activations Density 0.112%

    No Known Activations