© 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. 27648
    Prev
    Next
    INDEX
    Explanations

    properly functioning

    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
     vel
    -0.09
    è¿İ
    -0.09
    ä¸Ńç§ĭ
    -0.09
    ÙĪØŃ
    -0.09
    æĤ¬
    -0.09
    _SKIP
    -0.08
    è¿İæİ¥
    -0.08
    .frames
    -0.08
     harassment
    -0.08
    å°±è¿Ļæł·
    -0.08
    POSITIVE LOGITS
    å¥Ķèµ´
    0.10
    ä¸ŃåĽ½äºº
    0.10
    åıįè¿ĩæĿ¥
    0.09
    åĴĢ
    0.09
    Sit
    0.09
    éĢ¾
    0.08
    ä¸ij
    0.08
    å¿ĥæĥ³
    0.08
    çŃ¹
    0.08
    åį·
    0.08
    Activations Density 0.008%

    No Known Activations