© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20469
    Prev
    Next
    INDEX
    Explanations

    makes sense

    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
     weighed
    -0.10
    éĢŁçİĩ
    -0.10
    åģ¥åħ¨
    -0.10
     ЮÑĢ
    -0.10
     tam
    -0.09
    ä¹ł
    -0.09
    åħ¥ä½ı
    -0.09
    åIJĦè¡Į
    -0.09
    çŃ®
    -0.09
    Disclosure
    -0.09
    POSITIVE LOGITS
    åIJĪçIJĨ
    0.11
    éģĵçIJĨ
    0.11
    çļĦçIJĨçĶ±
    0.10
     reasons
    0.10
    çIJĨ
    0.10
    åIJĪçIJĨçļĦ
    0.10
     logical
    0.09
     reasonable
    0.09
    .logic
    0.09
    ä¸įçIJĨ
    0.09
    Activations Density 0.025%

    No Known Activations