© 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. 128953
    Prev
    Next
    INDEX
    Explanations

    change and improvement or techniques

    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
     emplois
    -1.02
    kový
    -1.01
     désormais
    -0.94
    ➗
    -0.94
     toată
    -0.92
    confirmPassword
    -0.90
     faptul
    -0.89
     déterminer
    -0.84
     Collagen
    -0.83
    ANTES
    -0.83
    POSITIVE LOGITS
     change
    1.50
     solutions
    1.43
     changes
    1.37
     interventions
    1.32
     improvement
    1.29
     by
    1.29
     corrective
    1.26
     techniques
    1.21
     methods
    1.20
     efforts
    1.19
    Activations Density 0.109%

    No Known Activations