© 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. 128637
    Prev
    Next
    INDEX
    Explanations

    papio

    np_acts-logits-general · gemini-2.5-flash-lite

    numerical sequences or digits in structured content, particularly in file references or data rows.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
     of
    -1.08
    sequently
    -1.05
     zimowa
    -1.02
    ——”
    -1.02
    dasarkan
    -1.01
     scorso
    -1.01
    無し
    -1.00
    ほら
    -1.00
     parecido
    -1.00
    あー
    -0.99
    POSITIVE LOGITS
     получили
    1.12
     only
    1.09
    と
    1.06
    <bos>
    1.02
    Demografie
    1.02
     مواليد
    1.02
     ま
    0.98
    6
    0.98
    是怎么回事
    0.98
     still
    0.97
    Activations Density 0.001%

    No Known Activations