© 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. 65330
    Prev
    Next
    INDEX
    Explanations

    saying goodbye

    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
    もあって
    -1.04
    儀式
    -0.97
    pekte
    -0.96
    中でも
    -0.93
     Gelder
    -0.92
    そのため
    -0.91
    アウト
    -0.90
    だが
    -0.90
    醍
    -0.89
    がありますが
    -0.89
    POSITIVE LOGITS
     goodbye
    1.29
    yled
    0.94
    accine
    0.90
     adieu
    0.90
     pájaro
    0.90
     geschafft
    0.89
    あぁ
    0.89
     davvero
    0.89
    おわりに
    0.89
     gemacht
    0.88
    Activations Density 0.023%

    No Known Activations