© 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. 15825
    Prev
    Next
    INDEX
    Explanations

    jax, MathJax, jaxrs

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

    The neuron fires on occurrences of the character sequence “jax” in tokens (e.g. in names like jax, jaxb, MathJax, lajax).

    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_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    </h3>
    -2.91
     That
    -2.44
     But
    -2.38
    ği
    -2.33
     ไม่
    -2.31
     nhưng
    -2.30
    '
    -2.27
     erwähnten
    -2.23
     ausdrück
    -2.22
    霂
    -2.20
    POSITIVE LOGITS
    s
    2.38
    糶
    2.20
    ጺ
    2.19
     concr
    2.19
     misi
    2.16
    (
    2.14
     topik
    2.11
    孩子们
    2.11
     perfe
    2.09
     privil
    2.09
    Activations Density 0.001%

    No Known Activations