© 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. 23126
    Prev
    Next
    INDEX
    Explanations

    junk, refuse, or bad quality

    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
    _frag
    -0.10
    çľĭçĹħ
    -0.10
    æĴ©
    -0.09
    çĤĺ
    -0.09
    zman
    -0.09
     Holt
    -0.09
    ç¿©
    -0.09
     Alive
    -0.09
    etz
    -0.08
    å¹ħ度
    -0.08
    POSITIVE LOGITS
    æ»ĵ
    0.10
    ä¿¡èµĸ
    0.10
    ä¸įå¿į
    0.09
     Champions
    0.09
     menuItem
    0.08
    stuff
    0.08
    apiro
    0.08
    ums
    0.08
    ä¸įåłª
    0.08
    宫廷
    0.08
    Activations Density 0.018%

    No Known Activations