© 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. 54332
    Prev
    Next
    INDEX
    Explanations

    series, linear feet, guide, resolution, study, linkage, .Linq

    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
    arb
    -0.10
    ãģ¬
    -0.09
    æµģè¡ĮçļĦ
    -0.09
    uploaded
    -0.09
    pawn
    -0.08
    æŀķ头
    -0.08
    ÑĥÑīе
    -0.08
    é¦ģ
    -0.08
    ç»Ī身
    -0.08
    伤å¿ĥ
    -0.08
    POSITIVE LOGITS
    ule
    0.09
    è¿Ļæī¹
    0.09
    falls
    0.09
    6
    0.09
     chall
    0.09
    0
    0.08
    4
    0.08
    5
    0.08
    8
    0.08
    9
    0.08
    Activations Density 0.063%

    No Known Activations