© 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. 34-GEMMASCOPE-RES-131K
    7. 107966
    Prev
    Next
    INDEX
    Explanations

    percentages, counts, frequencies

    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_34/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
    <0xBB>
    -0.83
    jā
    -0.75
    ━━━━
    -0.75
    tong
    -0.73
    Composition
    -0.71
     независи
    -0.69
    埴
    -0.68
     évident
    -0.68
    )}=
    -0.67
    GROW
    -0.67
    POSITIVE LOGITS
    Solu
    0.77
     obtained
    0.74
    取得
    0.73
     improvisation
    0.72
    orted
    0.72
    lehnt
    0.70
    าน
    0.68
     obtenu
    0.68
    lings
    0.68
     đa
    0.68
    Activations Density 0.032%

    No Known Activations