© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51010
    Prev
    Next
    INDEX
    Explanations

    made available

    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
    屿
    -0.08
    èį»
    -0.08
    wich
    -0.08
    æĹ¢æľī
    -0.08
    åIJĥæĥĬ
    -0.08
    åݦéŨ
    -0.08
    ä¼ij
    -0.08
    åļ·
    -0.08
    çij¾
    -0.08
    rosse
    -0.08
    POSITIVE LOGITS
    umbo
    0.10
    æĴ¬
    0.09
    å´ĸ
    0.08
    大çĽĺ
    0.08
    解æĶ¾æĢĿæĥ³
    0.08
    è´¼
    0.08
    ç´ł
    0.08
    abe
    0.08
    çģĮ
    0.08
    åĨľå®¶ä¹IJ
    0.08
    Activations Density 0.009%

    No Known Activations