© 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. 98445
    Prev
    Next
    INDEX
    Explanations

    Asian American identity

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

    This neuron strongly detects preposition tokens (“in,” “at,” “the”) that introduce locations or settings.

    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
    ├
    -0.87
     Zor
    -0.81
     Ov
    -0.80
     Thần
    -0.77
     Ấ
    -0.76
    Kentucky
    -0.74
    🏟
    -0.74
    NESSEE
    -0.73
     Thị
    -0.73
     Ε
    -0.73
    POSITIVE LOGITS
     Asian
    1.64
     Asians
    1.39
     asian
    1.38
    Asian
    1.36
     Chinatown
    1.30
     immigrant
    1.20
     Taiwanese
    1.10
     Flushing
    1.10
     immigrants
    1.08
    asian
    1.04
    Activations Density 0.069%

    No Known Activations