© 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. 103036
    Prev
    Next
    INDEX
    Explanations

    world followed by specific terms

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

    This neuron primarily detects occurrences of the word “World.”

    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
    TÜ
    -0.81
    kken
    -0.78
    Método
    -0.76
     основан
    -0.75
    ディ
    -0.75
     apparence
    -0.75
    Finnish
    -0.75
    linger
    -0.75
    Stk
    -0.74
     poudre
    -0.74
    POSITIVE LOGITS
    各地
    1.20
     world
    1.09
     Wide
    0.99
     famous
    0.95
     世界
    0.93
     largest
    0.93
    世界
    0.88
     világ
    0.88
    мир
    0.88
     worlds
    0.87
    Activations Density 0.017%

    No Known Activations