© 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. 64455
    Prev
    Next
    INDEX
    Explanations

    Earth originating; Q: queries; significant association

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

    This neuron activates on prominent section or table header terms—capitalized label words that mark categories or fields (e.g. “Archetype,” “Alchemical Correlation,” “Astrological Association,” etc.).

    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
    ּוֹ
    -2.64
    ִּ
    -2.28
    וּ
    -2.20
    ᾿
    -2.09
    ַּ
    -1.98
    ֶּ
    -1.95
    ֵּ
    -1.84
     séjours
    -1.78
     médec
    -1.77
     бампер
    -1.69
    POSITIVE LOGITS
     all
    1.85
     as
    1.84
     for
    1.84
     گیاه
    1.65
     first
    1.63
    ׇ
    1.57
     only
    1.56
     of
    1.56
     bringing
    1.53
     ویتامین
    1.52
    Activations Density 0.010%

    No Known Activations