© 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. 16294
    Prev
    Next
    INDEX
    Explanations

    java class paths

    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
     and
    -3.03
    をしました
    -1.77
     its
    -1.70
     escote
    -1.69
    やう
    -1.67
    づけ
    -1.65
    つづく
    -1.63
     not
    -1.63
    はな
    -1.63
     an
    -1.62
    POSITIVE LOGITS
    は
    2.22
    <bos>
    2.20
    では
    2.17
    などが
    2.08
    と
    1.99
    が
    1.93
    を
    1.92
    などを
    1.86
    で
    1.80
    もある
    1.74
    Activations Density 0.017%

    No Known Activations