© 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. 15815
    Prev
    Next
    INDEX
    Explanations

    technical terms and programming keywords

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

    This neuron detects uppercase metadata labels or section headings—i.e. all-caps tokens used as field names, headers, or tags in technical/configuration documents.

    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
     that
    -1.81
     (
    -1.79
    ”
    -1.72
     “
    -1.70
    ,”
    -1.62
     (“
    -1.61
    」、
    -1.56
    {~
    -1.55
     amarillas
    -1.52
    Although
    -1.49
    POSITIVE LOGITS
     we
    1.57
     eerste
    1.51
     první
    1.36
     اولین
    1.35
     tweede
    1.35
     procédé
    1.34
     nenhuma
    1.32
     masaje
    1.32
     necesita
    1.30
    ERVIEW
    1.30
    Activations Density 0.328%

    No Known Activations