© 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. 20270
    Prev
    Next
    INDEX
    Explanations

    maintain / remaining / continued state

    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
    phia
    -0.09
    鸣
    -0.08
    æ§
    -0.08
    ableView
    -0.08
    çŃĸ
    -0.08
    ãĤĴæ±ĤãĤģ
    -0.08
    Lim
    -0.08
    ocos
    -0.08
    woke
    -0.08
    .ft
    -0.08
    POSITIVE LOGITS
    ä¿ĿæĮģ
    0.31
    继ç»Ńä¿ĿæĮģ
    0.29
    ç»´æĮģ
    0.28
    ä¿ĿæĮģçĿĢ
    0.24
     maintain
    0.21
     maintaining
    0.21
    å§ĭç»Īä¿ĿæĮģ
    0.20
     maintained
    0.20
     ìľłì§Ģ
    0.20
     maintains
    0.19
    Activations Density 0.122%

    No Known Activations