© 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. 56470
    Prev
    Next
    INDEX
    Explanations

    can take or do

    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
    åĪĽæĸ°èĥ½åĬĽ
    -0.09
    æ´ĭ
    -0.09
    广éĺĶçļĦ
    -0.09
    strup
    -0.08
    翼
    -0.08
    è´¨çĸij
    -0.08
    macros
    -0.08
    æĪIJçĨŁçļĦ
    -0.08
    çħ§
    -0.08
    ç¾İå¾·
    -0.08
    POSITIVE LOGITS
    举æݪ
    0.11
    以åıĬåħ¶ä»ĸ
    0.11
    æݪæĸ½
    0.11
    éĩĩåıĸ
    0.11
    第ä¸ĢæŶéĹ´
    0.10
    æīĢèĥ½
    0.10
    åĸĦ
    0.10
    èĩªç§ģ
    0.09
    åģļå¾Ĺ
    0.09
    éĤĭ
    0.09
    Activations Density 0.055%

    No Known Activations