© 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. 22-GEMMASCOPE-RES-131K
    7. 32817
    Prev
    Next
    INDEX
    Explanations

    ““

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

    The neuron fires strongly on sentence-final markers—most prominently on periods (and the words immediately before them) indicating ends of sentences.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     that
    -0.98
     this
    -0.96
     to
    -0.93
    これが
    -0.91
     use
    -0.90
    はこの
    -0.88
     antisocial
    -0.81
     Mère
    -0.80
     This
    -0.79
    来店
    -0.79
    POSITIVE LOGITS
     “
    1.65
    ：“
    1.30
     "
    1.30
    他说
    1.30
     „
    1.25
    loggen
    1.20
    uteurs
    1.20
    Ꮠ
    1.14
    collant
    1.14
    formatie
    1.10
    Activations Density 0.016%

    No Known Activations