© 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. 11316
    Prev
    Next
    INDEX
    Explanations

    morning and day transitions

    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
    çĤ³
    -0.09
    ä½İ端
    -0.09
    è§ĴèIJ½
    -0.09
    ä¸įåħĭ
    -0.08
    æ¥Ķ
    -0.08
    èł¡
    -0.08
    请æĤ¨
    -0.08
    æ¿Ģæ´»
    -0.08
    娱ä¹IJ
    -0.08
    汪
    -0.08
    POSITIVE LOGITS
    第äºĮ天
    0.22
    ç¿Į
    0.21
    æ¸ħæĻ¨
    0.19
     tomorrow
    0.18
    æĺİ天
    0.18
    ä¸Ģ大æĹ©
    0.18
    æĹ©ä¸Ĭ
    0.17
     morning
    0.17
    æĹ©æĻ¨
    0.17
     Tomorrow
    0.15
    Activations Density 0.072%

    No Known Activations