© 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. 11058
    Prev
    Next
    INDEX
    Explanations

    tax exclusion, income, funding

    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°¸
    -0.10
    uctions
    -0.09
    ç¾
    -0.09
    é«ĺçŃī
    -0.09
    çī¹æ®ĬæĥħåĨµ
    -0.09
     Blond
    -0.08
    markets
    -0.08
    parity
    -0.08
    èģĮ
    -0.08
    羡
    -0.08
    POSITIVE LOGITS
    æīĢå¾Ĺ
    0.10
    çļĦç¾İ好
    0.10
    tu
    0.09
    æĶ¶åĽŀ
    0.09
    grounds
    0.09
    天ä¸Ĭ
    0.09
    è¡¥åħħ
    0.08
    交éĻħ
    0.08
     Volk
    0.08
    éĵ²
    0.08
    Activations Density 0.039%

    No Known Activations