© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38760
    Prev
    Next
    INDEX
    Explanations

    character names

    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
     each
    -1.41
     some
    -1.30
     themselves
    -1.27
    各自
    -1.05
     this
    -1.05
     these
    -1.04
    自己是
    -1.03
    لبة
    -1.02
    .《
    -1.01
    кас
    -1.00
    POSITIVE LOGITS
     who
    2.34
     whom
    2.06
     jego
    1.73
     его
    1.66
     his
    1.64
     whose
    1.59
     который
    1.55
     quien
    1.43
     שהוא
    1.41
    who
    1.40
    Activations Density 0.037%

    No Known Activations