© 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. 3271
    Prev
    Next
    INDEX
    Explanations

    numbers followed by units or counts

    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
    æ¦ľ
    -0.10
    满
    -0.09
    rets
    -0.09
     Sanity
    -0.09
    æ·¡åĮĸ
    -0.08
    æѦåĬŁ
    -0.08
    å¦Ĥä½ķåİ»
    -0.08
    豫
    -0.08
    average
    -0.08
    åħ¶ä¸Ńä¹ĭä¸Ģ
    -0.08
    POSITIVE LOGITS
    æĪijèĥ½
    0.11
    çļĦæŶéĹ´éĩĮ
    0.10
    ä¸įç»ıæĦı
    0.10
    éľı
    0.09
    å¹¿æ³ĽçļĦ
    0.09
    é¦ĸè½®
    0.09
    arta
    0.09
    olis
    0.08
    .exc
    0.08
    ä¸įäºĨ
    0.08
    Activations Density 0.020%

    No Known Activations