© 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. 15323
    Prev
    Next
    INDEX
    Explanations

    duplicate questions

    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
     because
    -1.43
     an
    -1.33
     their
    -1.27
     Taking
    -1.22
     from
    -1.18
     for
    -1.13
     anivers
    -1.13
     cleverly
    -1.09
     both
    -1.09
     mistakenly
    -1.09
    POSITIVE LOGITS
     tumultuous
    1.30
     áng
    1.28
    ことも
    1.26
     insanely
    1.26
    mainly
    1.22
     ottimo
    1.22
     чтобы
    1.22
    larg
    1.21
    シャレ
    1.20
     dismal
    1.20
    Activations Density 0.024%

    No Known Activations