© 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. 10-GEMMASCOPE-RES-131K
    7. 19268
    Prev
    Next
    INDEX
    Explanations

    Zhengzhou, Zhou Dynasty

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

    The neuron strongly activates on the Pinyin token “Zhou,” i.e. instances of “Zhou” in Chinese names, places, or dynastic terms.

    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
    ’
    -3.30
    鼍
    -2.66
     tratando
    -2.66
     повідом
    -2.48
     realizando
    -2.47
    久し
    -2.47
     colocando
    -2.45
     帅气
    -2.44
     présenter
    -2.41
     駅前
    -2.41
    POSITIVE LOGITS
    (
    3.38
    S
    2.86
    >
    2.70
    p
    2.59
    <em>
    2.50
    m
    2.50
                
    2.47
    .}(
    2.47
    l
    2.45
     impeccable
    2.44
    Activations Density 0.006%

    No Known Activations