© 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. 72330
    Prev
    Next
    INDEX
    Explanations

    student studies, VEGF expression, fatigue after

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

    words and phrases in Slavic languages, particularly recognizing Slavic linguistic patterns and text.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron activates on tokens containing non-ASCII letters or diacritics (e.g. accented or special characters in Slavic and other languages).

    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
     is
    -1.86
     would
    -1.81
     will
    -1.65
    TagHelpers
    -1.38
     COULD
    -1.36
     are
    -1.34
    Quién
    -1.32
     ถ้า
    -1.31
     exis
    -1.31
     were
    -1.30
    POSITIVE LOGITS
     only
    2.00
     a
    1.91
     any
    1.68
     both
    1.64
     even
    1.58
    gggg
    1.48
    鋮
    1.45
     people
    1.40
    ogues
    1.38
    eeeeeeee
    1.38
    Activations Density 0.008%

    No Known Activations