© 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. 48557
    Prev
    Next
    INDEX
    Explanations

    question words and particles

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

    The neuron responds to non-Latin (CJK) script tokens, especially Chinese/Japanese characters.

    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
    了呢
    -1.51
    的呢
    -1.26
     depending
    -1.26
    depending
    -1.05
    了啊
    -0.99
    でしょうね
    -0.88
     Nantes
    -0.87
    Depending
    -0.85
     呢
    -0.83
    ivan
    -0.81
    POSITIVE LOGITS
    吗
    2.98
    嗎
    2.70
     Does
    1.93
     does
    1.83
    ไหม
    1.79
     吗
    1.78
    Does
    1.78
     ли
    1.73
     آیا
    1.70
    么
    1.69
    Activations Density 0.007%

    No Known Activations