© 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. 57719
    Prev
    Next
    INDEX
    Explanations

    love with positive sentiment attachment

    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
    rek
    -0.10
    hev
    -0.09
    æīĢéľĢ
    -0.09
     Edmund
    -0.08
    åij¨å¹´
    -0.08
    路桥
    -0.08
     quo
    -0.08
    ses
    -0.08
    çīµæīĭ
    -0.08
    UDA
    -0.08
    POSITIVE LOGITS
    åIJ¬åĪ°
    0.11
    称ä¹ĭ
    0.10
     (?)
    0.10
    çľĭåĪ°
    0.10
    uet
    0.10
     ras
    0.09
    è§ģåĪ°
    0.09
    æľī人说
    0.09
    ulously
    0.09
    stras
    0.09
    Activations Density 0.023%

    No Known Activations