© 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. 34519
    Prev
    Next
    INDEX
    Explanations

    diagrams and technical illustrations

    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
    volent
    -0.10
    itaire
    -0.10
    è¿Ļ件
    -0.09
    ijo
    -0.09
    æĥħåĨµè¿Ľè¡Į
    -0.08
    ipient
    -0.08
    åİŁå§ĭ
    -0.08
    æĩĤäºĭ
    -0.08
    æİĢ
    -0.08
    说å®ŀè¯Ŀ
    -0.08
    POSITIVE LOGITS
    matic
    0.15
    æĻ°
    0.11
     Diagram
    0.11
     diagram
    0.11
    ä¸Ģå¹ħ
    0.11
     diagrams
    0.10
    sWith
    0.10
    atic
    0.10
    ly
    0.10
    仪
    0.09
    Activations Density 0.052%

    No Known Activations