© 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. 53188
    Prev
    Next
    INDEX
    Explanations

    make sure

    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
    åıĺå¾ĹæĽ´
    -0.09
    -scalable
    -0.09
    apl
    -0.08
    æļĹæļĹ
    -0.08
    æħ¢æħ¢åľ°
    -0.08
    aight
    -0.08
    æĹ¥å¸¸çĶŁæ´»
    -0.08
    agu
    -0.08
    acter
    -0.08
    enha
    -0.08
    POSITIVE LOGITS
    ç¡®ä¿Ŀ
    0.37
    確ä¿Ŀ
    0.30
     ensure
    0.27
    ä¿Ŀè¯ģ
    0.26
     ensures
    0.24
     ensuring
    0.24
     Ensure
    0.23
    Ensure
    0.23
    ensure
    0.23
    .ensure
    0.22
    Activations Density 0.088%

    No Known Activations