© 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. 22683
    Prev
    Next
    INDEX
    Explanations

    category headers

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

    The neuron is activated by standalone numeric tokens (e.g. list indices, version/build numbers, category labels, numeric literals).

    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
     like
    -2.94
     and
    -2.53
     only
    -2.52
     have
    -2.36
     on
    -2.30
     took
    -2.20
     had
    -2.20
     after
    -2.17
     give
    -2.16
     so
    -2.14
    POSITIVE LOGITS
     trám
    2.39
     probablement
    2.31
     accompagner
    2.19
     rafraî
    2.13
     souvent
    2.05
    だいぶ
    2.03
     refroid
    1.99
    usually
    1.98
    裥
    1.95
     magnific
    1.92
    Activations Density 0.002%

    No Known Activations