© 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. 129352
    Prev
    Next
    INDEX
    Explanations

    transition words introducing additions

    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
     however
    -1.27
     però
    -1.23
     either
    -1.18
     both
    -1.14
     pertanian
    -1.13
     gsx
    -1.09
     tumbuhan
    -1.09
     semacam
    -1.09
     allerdings
    -1.08
     piaggio
    -1.08
    POSITIVE LOGITS
    ,
    1.37
    何より
    1.30
     it
    1.29
    そもそも
    1.27
     wouldn
    1.20
     there
    1.13
     fact
    1.08
     Too
    1.07
     who
    1.05
    Too
    1.02
    Activations Density 0.013%

    No Known Activations