© 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. 33640
    Prev
    Next
    INDEX
    Explanations

    e.g., examples

    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
    ほうがいい
    -1.69
    んですよね
    -1.61
    toDouble
    -1.55
    んですね
    -1.50
    ところに
    -1.45
    ほうが
    -1.37
    toFloat
    -1.31
    方がいい
    -1.30
    んだろう
    -1.28
    わけで
    -1.27
    POSITIVE LOGITS
    や
    1.88
    を
    1.77
    <bos>
    1.76
    と
    1.61
    が
    1.50
    そして
    1.43
    は
    1.43
    では
    1.43
    これは
    1.41
    で
    1.40
    Activations Density 0.015%

    No Known Activations