© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 53783
    Prev
    Next
    INDEX
    Explanations

    absorb and digest information

    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
    æī§
    -0.12
    èĩ¼
    -0.10
    æŀ¶
    -0.10
    лÑĮ
    -0.09
    onen
    -0.09
    æĴ¬
    -0.09
    istar
    -0.09
     Hours
    -0.09
    uhl
    -0.09
    å¿Ĺ强
    -0.09
    POSITIVE LOGITS
    æ¶ĪåĮĸ
    0.14
    çļĦä¿¡æģ¯
    0.13
     information
    0.12
    æĽ´å¤ļä¿¡æģ¯
    0.11
    åIJ¸æĶ¶
    0.11
     инÑĦоÑĢмаÑĨиÑİ
    0.11
    èµĦ讯
    0.11
     digest
    0.10
    ä¿¡æģ¯
    0.10
     Informationen
    0.10
    Activations Density 0.063%

    No Known Activations