© 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. 54925
    Prev
    Next
    INDEX
    Explanations

    most + adjective/noun

    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
    hdl
    -0.10
    _mB
    -0.09
    缶
    -0.09
    pane
    -0.08
    irs
    -0.08
    _suite
    -0.08
    caff
    -0.08
     اعت
    -0.08
    ston
    -0.08
     Suite
    -0.08
    POSITIVE LOGITS
    GED
    0.10
    å¤ĩçĶ¨
    0.10
     Majesty
    0.09
    åIJĪæ³ķ
    0.09
    pcs
    0.09
    并äºİ
    0.09
    ç«ŀäºī对æīĭ
    0.09
     own
    0.09
    å¾ĹæĦı
    0.09
    迷人
    0.09
    Activations Density 0.086%

    No Known Activations