© 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. 34-GEMMASCOPE-RES-131K
    4. 98038
    Prev
    Next
    INDEX
    Explanations

    taken verbatim

    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_34/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
    égique
    -1.02
    归
    -0.73
    stå
    -0.72
    ध्यम
    -0.72
    aban
    -0.72
    Gian
    -0.72
     geschikt
    -0.71
    Contin
    -0.71
     旅游
    -0.71
     vielä
    -0.70
    POSITIVE LOGITS
     verbatim
    4.06
    verbatim
    2.91
     exact
    2.59
     exactly
    2.28
    exact
    2.13
    Exact
    2.05
     word
    2.03
     unchanged
    1.99
    batim
    1.98
     literal
    1.96
    Activations Density 0.104%

    No Known Activations