© 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. 31028
    Prev
    Next
    INDEX
    Explanations

    literal emphasis or exaggeration

    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
     realmente
    -1.45
     generell
    -1.26
     StatefulWidget
    -1.26
     vraiment
    -1.25
    really
    -1.21
    真正
    -1.17
    ౮
    -1.16
     general
    -1.13
     klima
    -1.12
    usiai
    -1.12
    POSITIVE LOGITS
     literally
    1.94
     cannot
    1.51
     hundreds
    1.45
     dozens
    1.38
     Literally
    1.34
     literalmente
    1.23
    Literally
    1.23
     thousands
    1.16
    literal
    1.16
     could
    1.13
    Activations Density 0.006%

    No Known Activations