© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 12810
    Prev
    Next
    INDEX
    Explanations

    browser compatibility issue

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

    The neuron strongly activates on numerical tokens—especially decimal numbers and percentages—in the text.

    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
     this
    -1.36
    França
    -1.26
     it
    -1.23
     navn
    -1.23
     теперь
    -1.22
     the
    -1.19
     etter
    -1.15
    Oto
    -1.14
    }}</
    -1.13
    ordnungen
    -1.13
    POSITIVE LOGITS
    only
    1.52
     فقط
    1.49
     only
    1.47
    replyreport
    1.46
    
    1.38
     illustratie
    1.34
     problemática
    1.33
    懨
    1.31
     unable
    1.30
     personalizado
    1.30
    Activations Density 0.081%

    No Known Activations