© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 48721
    Prev
    Next
    INDEX
    Explanations

    Jiang and Chiang names

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

    The neuron fires on occurrences of the substring “jiang,” i.e. Chinese names or place names containing “jiang.”

    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
    女性の
    -0.89
     Variante
    -0.87
    essore
    -0.84
     pprint
    -0.83
     laiton
    -0.82
    也不会
    -0.82
     abilities
    -0.81
     kiu
    -0.81
    şem
    -0.80
    Interess
    -0.80
    POSITIVE LOGITS
    welli
    0.94
     Our
    0.90
     Its
    0.90
     Resolución
    0.87
    accines
    0.87
    TeX
    0.83
     But
    0.83
     chết
    0.83
    hauer
    0.83
    CKS
    0.82
    Activations Density 0.004%

    No Known Activations