© 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. 50014
    Prev
    Next
    INDEX
    Explanations

    wrecked ship or car

    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
    overe
    -0.11
    ize
    -0.11
    æľĢ好ä¸įè¦ģ
    -0.10
    غ
    -0.09
    keit
    -0.09
     вÑģем
    -0.09
    纯æ´ģ
    -0.09
    ately
    -0.09
    æĮ¥
    -0.09
    éĽ
    -0.09
    POSITIVE LOGITS
     havoc
    0.13
    оп
    0.11
    æ»ĵ
    0.10
    alls
    0.09
    amp
    0.09
    äºĨä¸Ģ
    0.09
    ola
    0.09
    éĢħ
    0.08
    uder
    0.08
    ardy
    0.08
    Activations Density 0.007%

    No Known Activations