© 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. 38247
    Prev
    Next
    INDEX
    Explanations

    foreign discourse markers

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

    The neuron fires on conjugated verbs (inflected verb forms) across multiple languages.

    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
     nahezu
    -1.30
     äußerst
    -1.22
     those
    -1.16
     which
    -1.16
     innumerable
    -1.09
     niets
    -1.06
     alcuna
    -1.04
     these
    -1.03
    ಾಗ
    -1.02
     this
    -1.01
    POSITIVE LOGITS
     also
    2.58
     actually
    1.97
     indeed
    1.90
     therefore
    1.77
     however
    1.77
     moreover
    1.71
     furthermore
    1.59
     además
    1.57
     now
    1.56
     tiež
    1.55
    Activations Density 0.020%

    No Known Activations