© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 15481
    Prev
    Next
    INDEX
    Explanations

    Democrats calling out Republicans

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

    This neuron picks out the prominent words used in political news headlines—especially the capitalization and key party/policy terms that begin headline titles.

    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
    Also
    -1.64
     również
    -1.61
    There
    -1.52
    Even
    -1.52
     Mereka
    -1.51
     if
    -1.49
     or
    -1.48
    These
    -1.48
     jutaan
    -1.41
     вже
    -1.38
    POSITIVE LOGITS
     funcionando
    1.37
     장
    1.30
     разные
    1.30
     its
    1.30
     разных
    1.30
     navires
    1.27
     물
    1.25
    阝
    1.23
     chemins
    1.20
     éléments
    1.19
    Activations Density 0.094%

    No Known Activations