© 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. 611
    Prev
    Next
    INDEX
    Explanations

    named entities and associations

    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
     originalen
    -1.02
    begins
    -0.98
    全都
    -0.98
    worldwide
    -0.96
     ulei
    -0.96
    んだろう
    -0.95
     Reſ
    -0.95
     României
    -0.94
    contr
    -0.93
     Gratis
    -0.93
    POSITIVE LOGITS
     découvre
    1.03
     people
    0.98
     clinic
    0.97
     jsem
    0.96
    ScrollBar
    0.96
     where
    0.93
     bird
    0.93
     how
    0.93
     District
    0.92
     which
    0.91
    Activations Density 0.214%

    No Known Activations