© 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. 48416
    Prev
    Next
    INDEX
    Explanations

    where introducing confusion or problems

    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
    iolis
    -0.96
    囵
    -0.95
     suffisamment
    -0.92
    nothing
    -0.91
     kısa
    -0.90
     viciss
    -0.90
     vocale
    -0.90
    ったのは
    -0.90
     RESOURCE
    -0.89
     proste
    -0.89
    POSITIVE LOGITS
     begins
    1.78
     really
    1.77
     truly
    1.70
     begin
    1.59
     becomes
    1.45
     starts
    1.37
     began
    1.37
     differ
    1.36
     differences
    1.31
     comes
    1.30
    Activations Density 0.032%

    No Known Activations