© 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. 23636
    Prev
    Next
    INDEX
    Explanations

    suffix of words or names

    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
    åIJ¹
    -0.11
    èĩªåľ¨
    -0.11
    京津
    -0.10
    лÑij
    -0.10
    æĹłéļľç¢į
    -0.09
    ocop
    -0.09
    çģ¶
    -0.08
    å§ij
    -0.08
    ÑģÑĤав
    -0.08
    æ´ĭæ´ĭ
    -0.08
    POSITIVE LOGITS
    WARE
    0.11
    åĿª
    0.10
    èĤ©èĨĢ
    0.10
    å¤ıæĹ¥
    0.10
    ä¸į容
    0.10
    upy
    0.09
    èIJ¼
    0.09
    apel
    0.09
    æŃª
    0.08
    èĤ©
    0.08
    Activations Density 0.028%

    No Known Activations