© 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. 9459
    Prev
    Next
    INDEX
    Explanations

    Reverse-, -New, oida, taenia

    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
     swords
    -0.10
    è°Ľ
    -0.10
    èıľ
    -0.09
    鸾
    -0.09
    ä½ıæĪ·
    -0.09
    æĹº
    -0.09
    çĶ·ç¥ŀ
    -0.09
    çĪ¶åŃIJ
    -0.08
    æĩĴ
    -0.08
    èĭ±éĽĦ
    -0.08
    POSITIVE LOGITS
    åĬĽéĩı
    0.10
    大æī¹
    0.09
    ·æĸ°
    0.09
    ä¸ºçĽ®æłĩ
    0.09
    ANEL
    0.09
    leans
    0.09
    åĸī
    0.08
    æİ°
    0.08
    cerpt
    0.08
     Volk
    0.08
    Activations Density 0.064%

    No Known Activations