© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9964
    Prev
    Next
    INDEX
    Explanations

    stud or studied

    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
    ality
    -0.12
     иÑģкÑĥÑģ
    -0.09
     undue
    -0.08
    æĻģ
    -0.08
    umb
    -0.08
    ãģĹãĤĩãģĨ
    -0.08
     Carlson
    -0.08
    å·¥å§Ķ
    -0.08
    curring
    -0.08
     доволÑĮ
    -0.08
    POSITIVE LOGITS
    ãģĵãģĿ
    0.10
    ย
    0.09
    ķ
    0.09
    们çļĦ
    0.09
    iate
    0.09
    etail
    0.09
    å¯Ŀ室
    0.08
    BED
    0.08
    æŃ¤åŁºç¡Ģä¸Ĭ
    0.08
    æĹ¥è®°
    0.08
    Activations Density 0.008%

    No Known Activations