© 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. 38785
    Prev
    Next
    INDEX
    Explanations

    secret relationship, hidden character, secret song

    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
     ÐľÐ¾Ñģква
    -0.09
     stuck
    -0.08
     жиз
    -0.08
    ä¸ĭæĿ¥
    -0.08
    rights
    -0.08
    ä¸Ģå¹ħ
    -0.08
    èĻļ
    -0.08
    Paused
    -0.08
    è¦ĭãģŁ
    -0.08
    ByVersion
    -0.08
    POSITIVE LOGITS
    åģ·åģ·
    0.17
    ç§ĺå¯Ĩ
    0.17
    éļIJå½¢
    0.17
     hidden
    0.16
    éļIJèĹı
    0.16
    éļ±
    0.16
    éļIJèĶ½
    0.16
    éļIJ
    0.16
    çļĦç§ĺå¯Ĩ
    0.16
    æĤĦæĤĦ
    0.15
    Activations Density 0.084%

    No Known Activations