© 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. 70200
    Prev
    Next
    INDEX
    Explanations

    Middle East

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

    The neuron consistently activates on the token “Middle,” particularly when it’s part of the phrase “Middle East.”

    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
    '
    -3.70
     he
    -3.52
     we
    -3.02
    \
    -2.97
     "
    -2.84
    ’
    -2.52
    ,
    -2.41
     a
    -2.33
    *
    -2.33
     ‘
    -2.25
    POSITIVE LOGITS
    縯
    2.52
     bahwa
    2.48
     offerta
    2.42
    鄚
    2.36
     alcuna
    2.34
     ヶ
    2.31
     selepas
    2.27
     cami
    2.23
     amma
    2.23
    Especificación
    2.22
    Activations Density 0.004%

    No Known Activations