© 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. 16100
    Prev
    Next
    INDEX
    Explanations

    referenced earlier

    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
    ोर
    -1.02
     ー
    -1.02
    úrg
    -1.00
    anzug
    -0.97
    
    -0.97
     luscious
    -0.96
     aproximada
    -0.95
     BTU
    -0.95
     dreary
    -0.93
    موقع
    -0.92
    POSITIVE LOGITS
     mentioned
    2.44
     aforementioned
    2.19
     again
    2.09
     earlier
    2.09
     previously
    2.06
     above
    2.03
     described
    1.99
     discussed
    1.89
     as
    1.59
     that
    1.52
    Activations Density 0.044%

    No Known Activations