© 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. 48952
    Prev
    Next
    INDEX
    Explanations

    they followed by verb

    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
    æĪijåıªæĺ¯
    -0.11
    æĪij对
    -0.10
    å¿ł
    -0.10
     myself
    -0.09
    æĪijçŁ¥éģĵ
    -0.09
    æĪijåľ¨
    -0.09
    æĬĬæĪij
    -0.09
    åľ¨æĪijçļĦ
    -0.09
    让æĪij
    -0.09
     contrast
    -0.09
    POSITIVE LOGITS
    åħ±åIJĮ
    0.15
    åħ±åIJĮåĬªåĬĽ
    0.14
    åħ¨ä½ĵ
    0.13
    ä¸Ģèµ·åİ»
    0.13
    大家ä¸Ģèµ·
    0.13
    讨论
    0.13
    åħ±åIJĮä½ĵ
    0.12
    è¨İè«ĸ
    0.12
    ä¸Ģèµ·
    0.12
    èģĬ
    0.12
    Activations Density 0.149%

    No Known Activations