© 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. 59022
    Prev
    Next
    INDEX
    Explanations

    sharp enough / safe from shock / cut through

    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
    fs
    -0.09
    å½ĵå¹´
    -0.09
    å¦ĸ
    -0.09
    丣
    -0.08
    æ´Ĺè¡£
    -0.08
    usto
    -0.08
    ç§ijåѦçłĶ究
    -0.08
     Presidential
    -0.08
    ä¸Ģåı¥è¯Ŀ
    -0.08
     Conj
    -0.08
    POSITIVE LOGITS
     sharp
    0.13
    å°ĸ
    0.12
    Sharp
    0.12
    éĶIJ
    0.11
    sharp
    0.11
     Sharp
    0.11
    éĴĿ
    0.10
    ç²Ĺç³Ļ
    0.10
    éĶĭ
    0.10
    ç¢İçīĩ
    0.10
    Activations Density 0.050%

    No Known Activations