© 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. 57404
    Prev
    Next
    INDEX
    Explanations

    idiots, idyllic, id combat, id software

    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
    ero
    -0.10
    ickness
    -0.09
    yü
    -0.09
    è¿Ļä¹Ł
    -0.09
    çŃĶåį·
    -0.09
    ä¹Łä¸įæĥ³
    -0.08
    acular
    -0.08
    imbus
    -0.08
    请æķĻ
    -0.08
    izable
    -0.08
    POSITIVE LOGITS
    符
    0.10
    åĴĮå°ļ
    0.10
    ilik
    0.10
    èĭ«
    0.10
    ways
    0.09
    åĦĭ
    0.09
    å®ľ
    0.09
     sorte
    0.09
    ESS
    0.09
    =id
    0.09
    Activations Density 0.042%

    No Known Activations