© 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. 5121
    Prev
    Next
    INDEX
    Explanations

    across the

    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
    èĦ±
    -0.10
    åģıå·®
    -0.09
    èĢģåĮĸ
    -0.09
    å©¿
    -0.09
    å¾Ģ
    -0.08
    .Act
    -0.08
    åģı
    -0.08
    å¾·
    -0.08
    åĽ¤
    -0.08
    екÑģ
    -0.08
    POSITIVE LOGITS
    对éĿ¢
    0.10
    isto
    0.10
    ãĥĵãĥ«
    0.09
    allo
    0.09
    adena
    0.09
    bnb
    0.09
    UMB
    0.09
    绿水
    0.09
     accru
    0.09
    èĩ¨
    0.09
    Activations Density 0.016%

    No Known Activations