© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 101456
    Prev
    Next
    INDEX
    Explanations

    scent, association or alarm systems

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

    The neuron detects words naming fragrance “notes” or composition elements in perfume descriptions.

    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
     violento
    -1.43
     aumentando
    -1.38
     recommended
    -1.34
    立派
    -1.34
    鲧
    -1.33
     sogar
    -1.32
    cargador
    -1.30
     hermosas
    -1.30
     atractivos
    -1.30
     optar
    -1.27
    POSITIVE LOGITS
    RESUMEN
    1.32
    質な
    1.26
     as
    1.25
    rière
    1.19
     at
    1.18
    apache
    1.16
    acté
    1.16
    execSQL
    1.15
    renaline
    1.13
    DESCRIPCIÓN
    1.13
    Activations Density 0.007%

    No Known Activations