© 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. 6776
    Prev
    Next
    INDEX
    Explanations

    derivative, predicted, explicit

    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
    家å±ŀ
    -0.10
    æĴ¼
    -0.10
    ittle
    -0.09
    ç½Ł
    -0.09
     Stern
    -0.08
    auce
    -0.08
    èĬĤ
    -0.08
    ment
    -0.08
    lessly
    -0.08
    meric
    -0.08
    POSITIVE LOGITS
    ois
    0.12
    çļĦä¼ĺåĬ¿
    0.10
    ello
    0.10
    ELY
    0.10
    илÑĮ
    0.09
    arily
    0.09
    åı¨
    0.09
    ä¸Ģçķª
    0.09
    IOUS
    0.09
    æŁIJç§įç¨ĭ度
    0.09
    Activations Density 0.001%

    No Known Activations