© 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. 34649
    Prev
    Next
    INDEX
    Explanations

    naturally occurring

    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
    eken
    -0.10
    ä¸Ģå¼Ģå§ĭå°±
    -0.10
    æĴ©
    -0.10
    ä¸ĵçĶ¨
    -0.10
    .CG
    -0.09
    ibel
    -0.09
    ä¸Ģå¼Ģå§ĭ
    -0.09
    onium
    -0.09
    åħħåĪĨä½ĵçİ°
    -0.09
    ãģŀ
    -0.09
    POSITIVE LOGITS
     Trafford
    0.10
    èĩªåıij
    0.09
    unt
    0.09
    æĥħåĨµä¸ĭ
    0.09
     unt
    0.09
    -born
    0.09
    fully
    0.09
    çł´
    0.08
    ä¼Ĺ
    0.08
    æĦıä¹īä¸ĬçļĦ
    0.08
    Activations Density 0.090%

    No Known Activations