© 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. 36045
    Prev
    Next
    INDEX
    Explanations

    should have

    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
    æ´ĭæ´ĭ
    -0.10
    ika
    -0.09
    笺
    -0.08
    éĶ¦
    -0.08
    è´¢
    -0.08
    otti
    -0.08
    çŁ¿æ³īæ°´
    -0.08
    çĸ¤çĹķ
    -0.08
    粮油
    -0.08
    æĴ°åĨĻ
    -0.08
    POSITIVE LOGITS
    çĥ¦
    0.09
    éħIJ
    0.09
     consult
    0.09
    rough
    0.09
     Andrews
    0.08
    ä¹¡åľŁ
    0.08
    лиÑĨ
    0.08
     ero
    0.08
    kü
    0.08
    åįĩåѦ
    0.08
    Activations Density 0.033%

    No Known Activations