© 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. 34-GEMMASCOPE-RES-131K
    7. 783
    Prev
    Next
    INDEX
    Explanations

    sentence beginnings

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

    The neuron is chiefly activated by punctuation marks (e.g. commas, periods, semicolons, exclamation points).

    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
     hotell
    -1.13
    Ṇ
    -1.13
     dabar
    -1.06
     tyn
    -1.06
     datter
    -1.03
    可以说
    -0.98
    dă
    -0.98
     olie
    -0.97
    drawSprites
    -0.97
     elemen
    -0.95
    POSITIVE LOGITS
     you
    2.56
     if
    1.88
     they
    1.75
     don
    1.74
     beware
    1.69
     You
    1.49
     make
    1.48
     there
    1.48
     I
    1.47
     try
    1.47
    Activations Density 0.018%

    No Known Activations