© 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. 31641
    Prev
    Next
    INDEX
    Explanations

    tax cut, break, reduce, incentives, advantages

    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
    ewire
    -0.09
    é½
    -0.09
    opus
    -0.09
    ä¸į满
    -0.09
    就好
    -0.08
    rena
    -0.08
    ystore
    -0.08
    åijĨ
    -0.08
    åĪ¹è½¦
    -0.08
    æŀ³
    -0.08
    POSITIVE LOGITS
    æĥłåıĬ
    0.10
    åĩıè½»
    0.09
    为ä¼ģä¸ļ
    0.09
    éĻįä½İäºĨ
    0.09
    aster
    0.09
    éļ¾æ°ij
    0.08
    çĴĩ
    0.08
    .expires
    0.08
    .reduce
    0.08
    ç°Į
    0.08
    Activations Density 0.017%

    No Known Activations