© 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. 1565
    Prev
    Next
    INDEX
    Explanations

    calculations and trials

    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
    è´¨éĩıåĴĮ
    -0.17
    èĥ½åĬĽåĴĮ
    -0.17
    建设åĴĮ
    -0.14
    çIJĨ解åĴĮ
    -0.14
    æĬĢæľ¯åĴĮ
    -0.14
    管çIJĨåĴĮ
    -0.14
    æŶéĹ´åĴĮ
    -0.13
    ä¼ģä¸ļåĴĮ
    -0.13
    ç»ıéªĮåĴĮ
    -0.13
    并å°Ĩ
    -0.12
    POSITIVE LOGITS
     alike
    0.14
    缸ç»ĵåIJĪ
    0.13
    è¿Ļ两个
    0.11
    è¿Ļ两
    0.11
    çŃīæĥħåĨµ
    0.11
    äºĮèĢħ
    0.11
    åıĮéĩį
    0.11
     thereof
    0.11
    两èĢħ
    0.10
    çŃīéĹ®é¢ĺ
    0.10
    Activations Density 1.884%

    No Known Activations