© 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. 19687
    Prev
    Next
    INDEX
    Explanations

    lists, numbered items, or specific code elements

    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
    adal
    -0.09
    åħĭæĭī
    -0.09
     zespo
    -0.09
    üs
    -0.08
    ä½łçŁ¥éģĵ
    -0.08
    ächst
    -0.08
    ä½įæĸ¼
    -0.08
    奥æŀĹåĮ¹
    -0.08
    å¤Ħåľ¨
    -0.08
    .det
    -0.08
    POSITIVE LOGITS
    çļĦçIJĨçĶ±
    0.10
    çļĦéĤ£ç§į
    0.09
    åĩłä¹İ
    0.08
    Spoiler
    0.08
    æĪį
    0.08
    大æķ°æį®
    0.08
    ä¸įåħį
    0.08
    asi
    0.08
    æīĭ段
    0.08
    主线
    0.08
    Activations Density 0.237%

    No Known Activations