© 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. 22-GEMMASCOPE-RES-131K
    7. 70417
    Prev
    Next
    INDEX
    Explanations

    Latin America

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

    This neuron activates on tokens containing the substring “lat,” e.g. parts of words like latin, lathe, latch, or script tags including “Lat.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    vantaged
    -2.75
    ญี่ป
    -2.67
    CurtirCurtir
    -2.63
    傩
    -2.59
     ถ้า
    -2.58
     ส์
    -2.50
    ztes
    -2.48
    镠
    -2.39
    闿
    -2.34
    𖤣
    -2.34
    POSITIVE LOGITS
    \
    3.81
    ]
    2.81
     '
    2.73
    )
    2.56
    -
    2.50
    most
    2.45
    ar
    2.31
    your
    2.28
     begge
    2.23
    =
    2.17
    Activations Density 0.028%

    No Known Activations