© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52547
    Prev
    Next
    INDEX
    Explanations

    a: fiber, model, possible, tricky, potential

    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
    nable
    -0.09
    åºĩ
    -0.09
    åĸ§
    -0.09
    ExceptionHandler
    -0.09
    æĶ¾å¿ĥ
    -0.08
    ฯ
    -0.08
     Crossing
    -0.08
    utenberg
    -0.08
    nees
    -0.08
    åħħ足çļĦ
    -0.08
    POSITIVE LOGITS
    åIJĦ个æĸ¹éĿ¢
    0.10
    /of
    0.09
    æĹ¶æľŁ
    0.09
    çļĦ人çī©
    0.09
    ä»ħä¾Ľ
    0.09
    eks
    0.09
    åºĶçĶ¨åľºæĻ¯
    0.08
    ãĥĦ
    0.08
    éĺ¶å±Ĥ
    0.08
    æĹ¶æ®µ
    0.08
    Activations Density 0.292%

    No Known Activations