© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 55869
    Prev
    Next
    INDEX
    Explanations

    France and Paris

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¹īä¹Į
    -0.10
    /umd
    -0.10
    åķIJ
    -0.09
     NES
    -0.08
    otron
    -0.08
    å¼Ģåıij建设
    -0.08
    æ±ķ头
    -0.08
    locator
    -0.08
    è±Ĩ
    -0.08
    çķı
    -0.08
    POSITIVE LOGITS
    å·´é»İ
    0.23
    æ³ķåĽ½
    0.21
     France
    0.17
    France
    0.17
    ãĥķãĥ©ãĥ³ãĤ¹
    0.17
     Paris
    0.16
    Paris
    0.16
     French
    0.15
    French
    0.14
     france
    0.13
    Activations Density 0.617%

    No Known Activations