© 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. 29921
    Prev
    Next
    INDEX
    Explanations

    longstanding issues and problems

    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
    ä¸Ģè¾ĪåŃIJ
    -0.14
    æĸ¹éĿ¢çļĦ
    -0.10
    ç»ĪçĶŁ
    -0.10
    ç»Ī身
    -0.10
    æĸĩåĮĸ建设
    -0.10
    赡
    -0.09
    ноÑģÑĤей
    -0.09
    主ä¹īæĢĿæĥ³
    -0.09
    NST
    -0.09
    åĢĺ
    -0.09
    POSITIVE LOGITS
    ä¸Ģ个éĹ®é¢ĺ
    0.11
    å·²ä¹ħ
    0.11
    ago
    0.11
    çĽ¸ä¼ł
    0.10
    -established
    0.10
    åĨĻçĿĢ
    0.10
    æĤłä¹ħ
    0.09
    -old
    0.09
    ä¿ĿæĮģçĿĢ
    0.09
    ä¼ł
    0.09
    Activations Density 0.052%

    No Known Activations