© 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. 53525
    Prev
    Next
    INDEX
    Explanations

    doing well, doing right

    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
    arges
    -0.10
    çļĦä¿¡æģ¯
    -0.09
     Accountability
    -0.09
    ä¸Ģä½ĵ
    -0.09
    æīĢè°ĵçļĦ
    -0.09
    Intialized
    -0.09
    æĬ¹
    -0.09
    æļ´
    -0.09
    æĸ¡
    -0.09
     поÑģÑĤоÑıнно
    -0.09
    POSITIVE LOGITS
    åģļå¾Ĺ
    0.17
     done
    0.11
    æŃ£ç¡®
    0.11
    done
    0.10
    å¾Ĺ好
    0.10
    åĬŀ好
    0.09
    åģļæ³ķ
    0.09
    ç³
    0.09
    åģļ人
    0.09
    绺
    0.09
    Activations Density 0.132%

    No Known Activations