© 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. 3958
    Prev
    Next
    INDEX
    Explanations

    motion or direction

    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
    .ls
    -0.09
    å¸ĮæľĽèĩªå·±
    -0.09
    è¾ľè´Ł
    -0.09
    èIJ½åIJİ
    -0.09
    itting
    -0.09
    æĸĹå¿Ĺ
    -0.08
    quate
    -0.08
    terior
    -0.08
    èĪ¶
    -0.08
    ä¼ĺè´¨çļĦ
    -0.08
    POSITIVE LOGITS
    tw
    0.10
     requestId
    0.09
    svm
    0.09
    æŀģéĻIJ
    0.09
    iminal
    0.09
    zas
    0.09
     tw
    0.09
    ç¨ĭ度
    0.08
    åĬŀåħ¬
    0.08
    ä¸Ģçľĭ
    0.08
    Activations Density 0.004%

    No Known Activations