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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 39905
    Prev
    Next
    INDEX
    Explanations

    code comments and header files

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

    The neuron fires most strongly on comment‐related tokens (e.g. “//”, file‐extension “.h”, copyright symbols, author/license metadata), so it is detecting lines of code comments and header/license “boilerplate.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ngOnDestroy
    -0.91
     risotto
    -0.90
    epide
    -0.89
     dieselbe
    -0.88
     zahlreiche
    -0.88
     ayudará
    -0.88
     rumor
    -0.88
    見えます
    -0.87
     sämtliche
    -0.86
    屬於
    -0.86
    POSITIVE LOGITS
     ricord
    0.99
     olhos
    0.95
     vainqueur
    0.92
    toare
    0.91
    に戻る
    0.90
    corrhi
    0.89
     acquist
    0.87
     attiv
    0.87
     lác
    0.87
     dimentic
    0.86
    Activations Density 0.059%

    No Known Activations