© 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. 39583
    Prev
    Next
    INDEX
    Explanations

    look back at

    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
    åıijå±ķ空éĹ´
    -0.09
    ingo
    -0.09
    éĨĴäºĨ
    -0.08
    æ·¡æ°´
    -0.08
     reverse
    -0.08
    èµĸ以
    -0.08
    éĿłè¿ij
    -0.08
    èµ°åĩºåİ»
    -0.08
    åºĶ对
    -0.08
     âĸº
    -0.08
    POSITIVE LOGITS
    å¾Ģäºĭ
    0.13
    è¿ĩåİ»
    0.12
    è¾īçħĮ
    0.11
    è¿ĩåİ»çļĦ
    0.11
    å³¥
    0.11
    æĺ¯ä¸Ģ件
    0.10
    éģİåİ»
    0.09
    itech
    0.09
    è¿Ļåĩłå¹´
    0.09
    åįģå¤ļå¹´
    0.09
    Activations Density 0.021%

    No Known Activations