© 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. 18331
    Prev
    Next
    INDEX
    Explanations

    as planned, as expected, as usual, as normal

    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
    edImage
    -0.10
    ستر
    -0.09
    æ´Ĺ礼
    -0.09
    èģĶæĥ³
    -0.09
    ibrator
    -0.09
     bast
    -0.09
    é«ĺ度
    -0.08
    è¡Ĺ
    -0.08
     vp
    -0.08
    深深
    -0.08
    POSITIVE LOGITS
    lü
    0.09
    itra
    0.08
    ä¸įåĪ°ä½į
    0.08
    çĤ¬
    0.08
    APO
    0.08
    è´£
    0.08
    è¿ķ
    0.08
    åłª
    0.08
     Koch
    0.08
    SEMB
    0.08
    Activations Density 0.060%

    No Known Activations