© 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. 10-GEMMASCOPE-RES-131K
    7. 102721
    Prev
    Next
    INDEX
    Explanations

    numbered list item two

    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_10/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
     despite
    -2.31
     burgeoning
    -2.19
     eleven
    -2.19
     eight
    -2.16
     placing
    -2.13
     if
    -2.13
     thirteen
    -2.05
     frantic
    -2.03
     additional
    -2.03
     three
    -2.03
    POSITIVE LOGITS
     REALLY
    2.56
     exceedingly
    2.55
     VERY
    2.53
     unbelievably
    2.48
     decidedly
    2.48
     lauded
    2.41
     continually
    2.36
     ridiculously
    2.27
     !!!!
    2.25
     exceptionally
    2.22
    Activations Density 0.033%

    No Known Activations