© 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. 32979
    Prev
    Next
    INDEX
    Explanations

    contractions like didn't, haven't

    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
     Tahu
    -0.99
    klju
    -0.97
     Ikan
    -0.96
    particularly
    -0.95
     excès
    -0.93
     rodríguez
    -0.92
    阏
    -0.92
     TROP
    -0.92
     maniere
    -0.92
    UNIFIED
    -0.92
    POSITIVE LOGITS
     industrial
    0.93
     Used
    0.91
     PS
    0.85
     practical
    0.84
    arté
    0.84
     veía
    0.82
    LOAT
    0.81
     academic
    0.81
     all
    0.80
     H
    0.80
    Activations Density 0.040%

    No Known Activations