© 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. 30183
    Prev
    Next
    INDEX
    Explanations

    helping to achieve

    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
    åı¯ä»¥å¸®åĬ©
    -0.10
    ä¸įè§ģäºĨ
    -0.09
    æīĪ
    -0.09
    åı¯ä»¥è¯´
    -0.08
    ä¸Ģ个éĹ®é¢ĺ
    -0.08
    FORE
    -0.08
     Princip
    -0.08
    ä¾Ŀæŧæĺ¯
    -0.08
    建议
    -0.08
    listening
    -0.08
    POSITIVE LOGITS
    æłijç«ĭ
    0.12
     ÃŃch
    0.12
    fully
    0.10
    å¼ĢåĪĽ
    0.10
    å®ŀçİ°
    0.10
    çľĭåĩº
    0.10
    lessly
    0.10
    ç¡®ç«ĭ
    0.09
    roud
    0.09
    æħ°
    0.09
    Activations Density 0.171%

    No Known Activations