© 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. 86595
    Prev
    Next
    INDEX
    Explanations

    wind, sealing, styles

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

    This neuron detects occurrences of the root “seal” (including its English forms seal, sealed, sealing and the Chinese character 封).

    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
    ↵↵
    -3.22
    .]
    -2.06
    .\
    -1.98
    0
    -1.94
    s
    -1.94
     conhece
    -1.91
     όταν
    -1.76
     influenced
    -1.70
     decides
    -1.69
     perfecta
    -1.69
    POSITIVE LOGITS
    豨
    2.64
     spesifik
    2.48
     krim
    2.22
    𓆜
    2.16
     rigide
    2.16
     officielle
    2.13
    洐
    2.11
    ﮨ
    2.09
     makam
    2.08
     tidur
    2.06
    Activations Density 0.003%

    No Known Activations