© 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. 63306
    Prev
    Next
    INDEX
    Explanations

    Divergent, Borderlands, cities, Commander, Empires

    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
    rosso
    -0.09
    彩
    -0.09
    roman
    -0.09
    attles
    -0.09
    ç¥ŀè¯Ŀ
    -0.09
    RIC
    -0.08
    åıijæĮ¥
    -0.08
    èŁł
    -0.08
    agini
    -0.08
     inv
    -0.08
    POSITIVE LOGITS
    çļĦåŃ©åŃIJ
    0.09
    å¿«æīĭ
    0.08
    åŁİéĩĮ
    0.08
    æĺ¯ä¸Ģ件
    0.08
    åºķå±Ĥ
    0.08
    èµ°å¾Ĺ
    0.08
    _INF
    0.08
    çħ§
    0.08
    ä¹ĭåŁİ
    0.08
     civ
    0.07
    Activations Density 0.100%

    No Known Activations