© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 7019
    Prev
    Next
    INDEX
    Explanations

    numbers and percentages

    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
     frantic
    -1.02
     invigor
    -1.00
     because
    -0.94
     getting
    -0.94
     рік
    -0.92
     trying
    -0.91
     facilitating
    -0.90
     tailored
    -0.90
     had
    -0.90
     dazzling
    -0.89
    POSITIVE LOGITS
     apropri
    1.09
    mainly
    1.07
    relatively
    1.05
    larg
    1.01
    primarily
    0.99
    mostly
    0.98
    roughly
    0.98
    向了
    0.94
    commonly
    0.94
     magnific
    0.92
    Activations Density 0.010%

    No Known Activations