© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128442
    Prev
    Next
    INDEX
    Explanations

    text prompts, quotes, and code blocks

    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
     to
    -1.81
     in
    -1.43
    leyebilirsiniz
    -1.28
     veiks
    -1.27
    meuble
    -1.23
    to
    -1.21
     we
    -1.20
     (
    -1.19
    Encyklopedia
    -1.19
     papild
    -1.18
    POSITIVE LOGITS
    menü
    1.40
     piele
    1.38
    !}
    1.33
    Noter
    1.31
    ppure
    1.30
     étrang
    1.27
    Noticia
    1.25
    commented
    1.25
    Breite
    1.25
     poêle
    1.23
    Activations Density 0.001%

    No Known Activations