© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66917
    Prev
    Next
    INDEX
    Explanations

    code block identifiers

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

    The neuron is picking up on documentation‐style markup and keywords (e.g. Javadoc/RDoc comment delimiters, tags like {@…}, “brief,” “code,” asterisks, brackets, etc.) rather than actual code.

    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
     rése
    -1.71
     tişört
    -1.68
     réinitial
    -1.65
    aktivi
    -1.60
     coussin
    -1.58
     Ellos
    -1.58
     rafraî
    -1.56
     bluz
    -1.55
     dobré
    -1.55
    gorro
    -1.55
    POSITIVE LOGITS
     to
    2.09
     without
    1.96
     but
    1.93
     "
    1.91
     /
    1.86
     only
    1.86
     by
    1.67
     and
    1.67
     more
    1.66
     designed
    1.63
    Activations Density 0.041%

    No Known Activations