© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20566
    Prev
    Next
    INDEX
    Explanations

    best interest

    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
    ä¸Ģé¢Ĺ
    -0.10
    ä¸ĢæīĢ
    -0.09
    åı¯æĥ³
    -0.08
    Dou
    -0.08
    æłª
    -0.08
    审åĪ¤
    -0.07
    ä¸ĢæĬĬ
    -0.07
    读åIJİ
    -0.07
    ä¸ĢåĪ»
    -0.07
    çģĮ
    -0.07
    POSITIVE LOGITS
    åĪ©çĽĬ
    0.27
    çļĦåĪ©çĽĬ
    0.24
     interests
    0.21
    interest
    0.13
     interest
    0.13
     инÑĤеÑĢеÑģ
    0.12
     Interest
    0.12
     interés
    0.12
    åħ´è¶£
    0.11
    åIJĪæ³ķæĿĥçĽĬ
    0.11
    Activations Density 0.034%

    No Known Activations