© 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. 34-GEMMASCOPE-RES-131K
    7. 117427
    Prev
    Next
    INDEX
    Explanations

    translate, translations, text processing

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

    This neuron fires on references to Python’s string module—especially mentions of “string.punctuation” (and related use of `string` and its punctuation set).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Waterways
    -0.72
     Tron
    -0.70
     Zun
    -0.69
     Scholastic
    -0.69
     eo
    -0.68
     LSM
    -0.68
     Soko
    -0.66
    Pv
    -0.66
     RENE
    -0.65
    pdp
    -0.65
    POSITIVE LOGITS
    printable
    1.18
     translation
    0.92
     translational
    0.90
    translator
    0.89
    punctuation
    0.89
     printable
    0.86
    translate
    0.86
     translator
    0.81
    Translated
    0.80
    Più
    0.80
    Activations Density 0.024%

    No Known Activations