© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 82509
    Prev
    Next
    INDEX
    Explanations

    german language constructs

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

    The neuron fires on German determiners—especially the definite articles (e.g. “der,” “die,” “das”) and related forms (e.g. “zur,” “diese”).

    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
     Trennung
    -0.86
     réputation
    -0.86
    美麗
    -0.82
     Freundschaft
    -0.82
     alibi
    -0.81
     postulated
    -0.81
    許多
    -0.80
    servicio
    -0.79
     イメージ
    -0.79
     Verantwortung
    -0.78
    POSITIVE LOGITS
     heo
    0.85
     Durchschnitt
    0.83
     adam
    0.80
    chieht
    0.79
    颇
    0.78
     Unglück
    0.77
    味わ
    0.77
    hose
    0.77
     contamination
    0.76
    sial
    0.74
    Activations Density 0.006%

    No Known Activations