© 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. 30114
    Prev
    Next
    INDEX
    Explanations

    work or tasks

    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
    èµłéĢģ
    -0.09
    çĶ¨äºº
    -0.09
    çĶ¨å·¥
    -0.09
    滥
    -0.09
    æĿŀ
    -0.09
    æŶéĹ´åĨħ
    -0.09
    缮åīįåĽ½åĨħ
    -0.08
    èŀ
    -0.08
     Laud
    -0.08
    arian
    -0.08
    POSITIVE LOGITS
    ä»»åĬ¡
    0.09
    纺
    0.09
    飧
    0.08
     tess
    0.08
    ä¸ĭä¸Ģ代
    0.08
    ãĤ¹ãĥĿãĥĥãĥĪ
    0.08
    åºķ
    0.08
    åĨĽå·¥
    0.08
    enko
    0.08
     tasks
    0.08
    Activations Density 0.207%

    No Known Activations