© 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. 37906
    Prev
    Next
    INDEX
    Explanations

    provided by

    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
    åľ¨äº¬
    -0.09
    åľ¨å¹¿å·ŀ
    -0.09
    oston
    -0.09
    å¿ĥæĦı
    -0.09
    åľ¨åĮĹ京
    -0.09
    ró
    -0.08
     ÑģовеÑĢÑĪен
    -0.08
    halt
    -0.08
    æĦı
    -0.08
    iani
    -0.08
    POSITIVE LOGITS
    第ä¸īæĸ¹
    0.11
    éĶĢ
    0.10
    .Animation
    0.09
    æĹłç©·
    0.09
    .SDK
    0.08
    _CAT
    0.08
    ä¸ĩæĪ·
    0.08
    RL
    0.08
    zell
    0.08
    çħ
    0.08
    Activations Density 0.040%

    No Known Activations