© 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. 48057
    Prev
    Next
    INDEX
    Explanations

    place names followed by distinguishing words

    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
    两èĢħ
    -0.09
    åĴĮå®ĮåĸĦ
    -0.09
    alien
    -0.09
    èĪĽ
    -0.08
    çĮĸ
    -0.08
    ASY
    -0.08
    åįłé¢Ĩ
    -0.08
    æºŁ
    -0.08
    åķ®
    -0.08
    ä½ĵåĨħ
    -0.08
    POSITIVE LOGITS
    äºĮ代
    0.10
    ä¸ĸ家
    0.09
     whom
    0.08
    é¡¹çĽ®çļĦ
    0.08
    ä¸Ģç¬Ķ
    0.08
    stime
    0.08
    代表
    0.08
    ä½Ĩä¸įéĻIJ
    0.08
     fame
    0.08
    èĦį
    0.08
    Activations Density 0.155%

    No Known Activations