© 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. 9576
    Prev
    Next
    INDEX
    Explanations

    desirable, preferred, recommended, advice

    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
    åħĢ
    -0.10
    å°ģ
    -0.10
    å¾Īæ·±
    -0.09
    åįģæĿ¡
    -0.09
    çĪµ
    -0.09
    身åŃIJ
    -0.09
    ä¸İä¸ŃåĽ½
    -0.09
     пÑĢож
    -0.08
    è¡Ķ
    -0.08
    ç»Ħç»ĩå®ŀæĸ½
    -0.08
    POSITIVE LOGITS
    é¦ĸéĢī
    0.10
     preferable
    0.10
    æľ
    0.09
    åĸĩåıŃ
    0.09
    é¼ĵåĬ±
    0.09
    æıIJåĢ¡
    0.09
    åıij声
    0.08
     erw
    0.08
    è´£
    0.08
     desirable
    0.08
    Activations Density 0.092%

    No Known Activations