© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9137
    Prev
    Next
    INDEX
    Explanations

    a word

    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
    说æĪij
    -0.10
    ä¸Ĭå²Ĺ
    -0.09
    许
    -0.09
    èļ¤
    -0.09
    æĪIJé¾Ļ
    -0.09
    æĺ¯çľŁçļĦ
    -0.09
    æĹ¥åŃIJ
    -0.09
    è¯ļ
    -0.09
    zos
    -0.09
    èĵĬ
    -0.08
    POSITIVE LOGITS
    çļĦä¸Ģéĥ¨åĪĨ
    0.19
    çļĦä¸Ģåijĺ
    0.17
    çļĦä¸Ģç§į
    0.13
    çļĦä¸Ģ大
    0.12
    æīĭ段
    0.11
     помоÑī
    0.10
    åĶ¯ä¸Ģ
    0.10
    åIJĪæł¼
    0.10
    çļĦä¸Ģ项
    0.10
    Integral
    0.09
    Activations Density 0.332%

    No Known Activations