© 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. 57396
    Prev
    Next
    INDEX
    Explanations

    finding their way into

    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
    群
    -0.10
    é©¿
    -0.09
    åī§
    -0.09
    enza
    -0.08
    棣
    -0.08
    iesel
    -0.08
    åij½
    -0.08
     inters
    -0.08
    ÑıÑĤ
    -0.08
    ç¬Ļ
    -0.08
    POSITIVE LOGITS
    åĩºçİ°åľ¨
    0.13
    èµ°ä¸Ĭ
    0.10
    æµģéĢļ
    0.10
    æĿ¥åĪ°
    0.10
    è¿ĽäºĨ
    0.10
    è¿Ľåħ¥åĪ°
    0.10
    æĿ¥åĪ°äºĨ
    0.10
    移åĪ°
    0.09
    è¿Ľåħ¥äºĨ
    0.09
    æµģåħ¥
    0.09
    Activations Density 0.054%

    No Known Activations