© 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. 61280
    Prev
    Next
    INDEX
    Explanations

    leaders, soldiers, learners, kids, agents, researchers

    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
    éļľ
    -0.10
    çłĶ究éĻ¢
    -0.09
    /stream
    -0.09
    Blocking
    -0.09
    å¤ĸæĿ¥
    -0.08
    忽
    -0.08
    åĪ¥äºº
    -0.08
     Instructor
    -0.08
    åıĭæĥħ
    -0.08
    å¤ĸèµĦ
    -0.08
    POSITIVE LOGITS
    æĩĤäºĭ
    0.10
    äºĨä¸Ģæī¹
    0.10
    ç´łåħ»
    0.10
    æłĭ
    0.10
     Foley
    0.09
    综åIJĪç´łè´¨
    0.09
    åħ¨èĥ½
    0.09
    åĽŀ头
    0.09
    åħ»æĪIJ
    0.09
     capable
    0.08
    Activations Density 0.047%

    No Known Activations