© 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. 25928
    Prev
    Next
    INDEX
    Explanations

    two or both

    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
     many
    -1.03
     that
    -1.03
     when
    -1.00
     in
    -0.97
     even
    -0.97
     while
    -0.97
     generally
    -0.96
     by
    -0.95
     during
    -0.95
    的一个
    -0.91
    POSITIVE LOGITS
     respectively
    2.13
    respectively
    2.05
     beiden
    1.98
     respectivement
    1.96
    分别
    1.88
     δύο
    1.78
    分別
    1.76
     two
    1.74
     две
    1.73
     både
    1.73
    Activations Density 0.222%

    No Known Activations