© 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. 54322
    Prev
    Next
    INDEX
    Explanations

    learning and teaching

    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
    失信
    -0.10
    ä¼ij
    -0.09
    è±¢
    -0.09
    unsch
    -0.08
    yu
    -0.08
     gross
    -0.08
    åĪĨæķ°çº¿
    -0.08
    辦
    -0.08
    æĮģèĤ¡
    -0.08
    çĹ£
    -0.08
    POSITIVE LOGITS
    æĢ¥éĢŁ
    0.10
    åı¯éĢļè¿ĩ
    0.09
    cheme
    0.09
    ä¹ĭå¤ĸ
    0.08
    uite
    0.08
    ä¸Ģéģį
    0.08
    è§Ī
    0.08
    æģķ
    0.08
    çļĦä¸ľè¥¿
    0.08
     alone
    0.08
    Activations Density 0.086%

    No Known Activations