© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 192131
    Prev
    Next
    INDEX
    Explanations

    Gestión, garbage, retort, gratitude

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

    The neuron strongly responds to content-word tokens used as section headings or topic labels (e.g., Spanish headings like "Gestión").

    oai_token-act-pair · gpt-5-mini
    Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     Ade
    0.42
    ade
    0.41
     свой
    0.39
     Tagen
    0.39
    adino
    0.39
    olines
    0.38
    骋
    0.38
    victoria
    0.38
     avsl
    0.38
    adanam
    0.38
    POSITIVE LOGITS
    ٭
    0.43
     dimensionality
    0.43
     CTR
    0.41
     Breathing
    0.41
     DA
    0.40
     picture
    0.39
     graphique
    0.38
     CNC
    0.38
     Doppler
    0.38
     plz
    0.38
    Activations Density 0.001%

    No Known Activations