© 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. 17662
    Prev
    Next
    INDEX
    Explanations

    Michigan, Lansing locations

    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
    港
    -0.10
    æ±ķ头
    -0.10
    æµİå®ģ
    -0.09
    塱
    -0.09
    åľ¨é¦Ļ港
    -0.09
    ument
    -0.09
    åĨ¼
    -0.09
    é¦Ļ港
    -0.09
    å²ŃåįĹ
    -0.09
    åľ¨å¹¿å·ŀ
    -0.09
    POSITIVE LOGITS
     Michigan
    0.20
     Detroit
    0.19
    Michigan
    0.18
    Detroit
    0.17
    æ»ģ
    0.15
     Wolver
    0.13
    èĬľæ¹ĸ
    0.12
    æ¯ħ
    0.10
    èĬ¬åħ°
    0.10
    æĿŁ
    0.10
    Activations Density 0.238%

    No Known Activations