© 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. 82382
    Prev
    Next
    INDEX
    Explanations

    NBA guards and players

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

    The neuron strongly activates on punctuation separators—especially commas—marking clause or list boundaries.

    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
    senior
    -0.91
     kontakta
    -0.85
     فاط
    -0.84
     diskusi
    -0.81
    ◯
    -0.80
     lemma
    -0.77
     poche
    -0.76
     amerik
    -0.75
     krep
    -0.75
     kvadrat
    -0.74
    POSITIVE LOGITS
    joven
    0.77
    退
    0.71
     Feuchtigkeit
    0.70
     Kenyon
    0.68
    修行
    0.68
     Dwight
    0.67
     Bildung
    0.66
     Hedge
    0.66
    nginx
    0.66
    Pe
    0.65
    Activations Density 0.043%

    No Known Activations