© 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. 30579
    Prev
    Next
    INDEX
    Explanations

    table of contents

    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
    ä¸įèĥ½åĨį
    -0.11
    è¯Ŀ说
    -0.11
    家éķ·
    -0.10
    ä¸ĭå·´
    -0.10
    ITU
    -0.10
    åĪĹ为
    -0.10
    家ä½ı
    -0.10
    åijĪçİ°
    -0.09
    ipi
    -0.09
    麻çĹ¹
    -0.09
    POSITIVE LOGITS
    upp
    0.08
    urer
    0.08
     undone
    0.08
    大èĥĨ
    0.08
    å©´
    0.08
    golden
    0.08
    ropp
    0.08
    èĭĴ
    0.08
     interf
    0.08
    chn
    0.08
    Activations Density 0.004%

    No Known Activations