© 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. 19612
    Prev
    Next
    INDEX
    Explanations

    family or genus

    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
    顺
    -0.12
    icism
    -0.09
    对åѦçĶŁ
    -0.09
    éłĨ
    -0.09
    ÑĤвоÑĢ
    -0.09
    èIJĮ
    -0.08
    à¹Ģà¸ī
    -0.08
    ire
    -0.08
    åıijèĤ²
    -0.08
    awa
    -0.08
    POSITIVE LOGITS
    大äºĭ
    0.11
    é»ĦåŁĶ
    0.10
    ä¼¼çļĦ
    0.09
     Hollow
    0.09
    å±ŀ
    0.09
     family
    0.09
    åķ¤
    0.09
    OLLOW
    0.09
    åı£è¢ĭ
    0.08
    æĻ°
    0.08
    Activations Density 0.292%

    No Known Activations