© 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. 49626
    Prev
    Next
    INDEX
    Explanations

    unintended mistakes or errors

    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
    åºĶæĶ¶
    -0.09
     isp
    -0.08
     obsess
    -0.08
    æĹĸ
    -0.08
    èĢĻ
    -0.08
    ÑĨин
    -0.08
    _UNLOCK
    -0.08
     Booker
    -0.08
     raj
    -0.08
    äºĭä¸ļ
    -0.08
    POSITIVE LOGITS
    æĻºåĬĽ
    0.10
    åįķ纯çļĦ
    0.10
    ä¸ī代
    0.10
    失误
    0.09
    åįķ纯
    0.09
    protocols
    0.09
    ä¿ĿåĪ©
    0.09
     unintention
    0.09
    æĦıè¯Ĩ
    0.08
    åĩºäºİ
    0.08
    Activations Density 0.046%

    No Known Activations