© 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. 10-GEMMASCOPE-RES-131K
    7. 67387
    Prev
    Next
    INDEX
    Explanations

    marked or pronounced degree

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on occurrences of the adjective “pronounced,” i.e. when a marked or emphasized difference is being described.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     testé
    -1.66
     sélectionné
    -1.59
    迓
    -1.55
     okno
    -1.52
     souligné
    -1.52
     increíble
    -1.50
    떳
    -1.48
     otras
    -1.46
     précisé
    -1.46
     muchos
    -1.45
    POSITIVE LOGITS
    '
    1.41
     marcadas
    1.34
    ’
    1.33
     tiff
    1.31
     начале
    1.30
    торт
    1.28
    ）；
    1.27
    äksi
    1.25
     which
    1.25
     waarmee
    1.23
    Activations Density 0.037%

    No Known Activations