© 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. 9783
    Prev
    Next
    INDEX
    Explanations

    toss with

    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
    çݰ代社ä¼ļ
    -0.10
    举è¡ĮçļĦ
    -0.09
    åģļçļĦäºĭæĥħ
    -0.09
    åĽ½å®¶çº§
    -0.08
    æĸ°ä¸Ģ代
    -0.08
    olle
    -0.08
     Bip
    -0.08
    æģķ
    -0.08
    å¾Ĺèµ·
    -0.08
    otify
    -0.08
    POSITIVE LOGITS
    CLA
    0.11
    -spin
    0.09
    shint
    0.08
    åľ¨ä¸Ģèµ·
    0.08
    oth
    0.08
     camer
    0.08
    _periods
    0.08
    åħ¶å®ĥéĹ®é¢ĺ
    0.08
    ghi
    0.08
    eries
    0.08
    Activations Density 0.025%

    No Known Activations