© 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. 7629
    Prev
    Next
    INDEX
    Explanations

    honeymoon or couple

    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
    èģª
    -0.09
    çĸ¤
    -0.09
     Baldwin
    -0.08
    fy
    -0.08
    çĩ
    -0.08
    严éĩį
    -0.08
    èĬ¹
    -0.08
    è¥ŀ
    -0.08
    èĭ±æīį
    -0.08
    麾
    -0.08
    POSITIVE LOGITS
    æģĭçĪ±
    0.19
    çĪ±æĥħ
    0.19
    缸çĪ±
    0.18
    äºĮ人
    0.17
    夫妻
    0.17
    æģĭæĥħ
    0.17
    çļĦçĪ±æĥħ
    0.17
    è°ĪæģĭçĪ±
    0.16
    æĥħä¾£
    0.16
    åıĮæĸ¹
    0.16
    Activations Density 0.112%

    No Known Activations