© 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. 26260
    Prev
    Next
    INDEX
    Explanations

    mime multipart email parsing

    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
     their
    -1.48
     this
    -1.45
     on
    -1.38
    0
    -1.34
     will
    -1.30
     also
    -1.30
     see
    -1.23
     the
    -1.23
     we
    -1.21
     by
    -1.19
    POSITIVE LOGITS
    很赞
    1.34
    Kontra
    1.30
     unden
    1.27
     rega
    1.27
     inconce
    1.26
    黩
    1.26
    욤
    1.24
    Authenticate
    1.22
    解决了
    1.21
    がたくさん
    1.20
    Activations Density 0.008%

    No Known Activations