© 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. 25625
    Prev
    Next
    INDEX
    Explanations

    not working or does not solve

    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
     patienter
    -1.23
    OOTDTY
    -1.22
    ilige
    -1.20
     verbe
    -1.16
     berge
    -1.14
     geprü
    -1.13
     ܙ
    -1.13
     oublier
    -1.11
     occuper
    -1.10
     faciliter
    -1.08
    POSITIVE LOGITS
    řízen
    1.16
     hasta
    0.96
    िल्म
    0.94
     might
    0.91
    vía
    0.89
    ਊ
    0.89
    できました
    0.89
    řízení
    0.88
     烏
    0.88
     it
    0.86
    Activations Density 0.067%

    No Known Activations