© 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. 5515
    Prev
    Next
    INDEX
    Explanations

    problems and recalls

    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
     Tough
    -0.10
    ummer
    -0.10
    æĪIJ交
    -0.09
    æĶ¹éĿ©å¼ĢæĶ¾
    -0.09
    æİ¢
    -0.09
    éŀ
    -0.09
    åįıåĬŀ
    -0.09
    eras
    -0.09
     Ñĥдален
    -0.09
    uide
    -0.08
    POSITIVE LOGITS
    è°ĪæģĭçĪ±
    0.10
    æĻ®éģį
    0.10
    导èĩ´
    0.10
    缺éĻ·
    0.10
    è¿Ļ个éĹ®é¢ĺ
    0.10
    è¿Ļæī¹
    0.10
    ç¦Ģ
    0.09
    å®īåħ¨æĢ§
    0.09
    è¶ĭ
    0.09
    è´¨éĩıéĹ®é¢ĺ
    0.09
    Activations Density 0.066%

    No Known Activations