© 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. 38272
    Prev
    Next
    INDEX
    Explanations

    can be

    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
     however
    -0.16
    however
    -0.13
    çĦ¶èĢĮ
    -0.12
    æĸĩåĮĸåĴĮ
    -0.12
    çIJĨ解åĴĮ
    -0.12
     allerdings
    -0.11
    ãģĿãģĹãģ¦
    -0.10
    并ä¸Ķ
    -0.10
    èĥ½åĬĽåĴĮ
    -0.10
     jedoch
    -0.10
    POSITIVE LOGITS
    为ä¹ĭ
    0.12
    缸åºĶçļĦ
    0.11
     ÙĦØ°ÙĦÙĥ
    0.11
     accordingly
    0.11
    rog
    0.10
    éļıä¹ĭ
    0.10
    çĶ±æѤ
    0.10
    缸åºĶ
    0.10
     darn
    0.10
    ä»İä¸Ń
    0.09
    Activations Density 0.806%

    No Known Activations