© 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. 81540
    Prev
    Next
    INDEX
    Explanations

    the ground or earth

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

    The neuron is highly sensitive to common English function words—especially short articles and prepositions like “the,” “a,” “in,” and “from.”

    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
     UID
    -0.68
    ديو
    -0.66
     Ingredient
    -0.66
    erein
    -0.66
     lont
    -0.66
    unn
    -0.66
    piti
    -0.66
    avs
    -0.66
     Führung
    -0.65
    restes
    -0.65
    POSITIVE LOGITS
     ground
    4.38
     earth
    3.17
    ground
    2.95
    Ground
    2.67
    地面
    2.55
     Earth
    2.53
     Ground
    2.47
    GROUND
    2.39
     GROUND
    2.36
     زمین
    2.19
    Activations Density 0.042%

    No Known Activations