© 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. 73188
    Prev
    Next
    INDEX
    Explanations

    smooth, clean surfaces

    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
    -1.19
     then
    -1.18
     пять
    -1.17
     festivities
    -1.06
     Obtén
    -1.04
     blackish
    -1.03
    AuthGuard
    -1.02
    łaści
    -1.01
     peo
    -1.01
     hivyo
    -1.01
    POSITIVE LOGITS
     clean
    1.93
     smooth
    1.66
     flawless
    1.28
    smooth
    1.27
    clean
    1.26
     cleans
    1.20
     perfect
    1.19
     uninterrupted
    1.19
    Clean
    1.16
     cleaner
    1.16
    Activations Density 0.052%

    No Known Activations