© 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. 18084
    Prev
    Next
    INDEX
    Explanations

    as ... as

    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
    大巴
    -0.10
    临
    -0.10
    姥
    -0.09
    ä¸įæĺĵ
    -0.08
    滥
    -0.08
    çľ¼è§Ĵ
    -0.08
    没éĶĻ
    -0.08
    ilik
    -0.08
    fest
    -0.08
    åĴ¸
    -0.08
    POSITIVE LOGITS
    æĪij们认为
    0.12
     nor
    0.12
    nor
    0.12
    èĢĮæĺ¯
    0.11
    ä»İåīį
    0.10
    arf
    0.10
    ä¹Łä¸įæĥ³
    0.10
    åºĶæĶ¶
    0.09
     anymore
    0.09
    QUI
    0.09
    Activations Density 0.012%

    No Known Activations