© 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. 48768
    Prev
    Next
    INDEX
    Explanations

    watching movies or TV shows

    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
    çIJ´
    -0.09
    åĩºæĿ¥äºĨ
    -0.09
    礼
    -0.09
     Spect
    -0.09
    èĬĿ
    -0.08
    ä½ĵ彩
    -0.08
    çĤ®
    -0.08
     اÙĦÙħÙĨا
    -0.08
    åĮĹ大
    -0.08
    çĽIJ
    -0.08
    POSITIVE LOGITS
     thụ
    0.09
    çŃīåĪ°
    0.09
    GINE
    0.09
    åĽŀåIJĪ
    0.09
    .setHorizontalAlignment
    0.09
    ovsky
    0.08
    é¢ĨæĤŁ
    0.08
    姹
    0.08
     geh
    0.08
     thi
    0.08
    Activations Density 0.056%

    No Known Activations