© 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. 5622
    Prev
    Next
    INDEX
    Explanations

    that it

    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¸Ģæĸ¹
    -0.09
    ours
    -0.09
    è¿ĻéĩĮçļĦ
    -0.08
    éºĵ
    -0.08
    /vendors
    -0.08
    æĹ¶ä»£çļĦ
    -0.08
    æĬĬè¿ĻäºĽ
    -0.08
    åħ¬åı¸çļĦ
    -0.08
    ä¸ĭçļĦ
    -0.07
    ળ
    -0.07
    POSITIVE LOGITS
    å®ĥ
    0.36
    å®ĥåı¯ä»¥
    0.29
    å®ĥæĺ¯
    0.29
    å®ĥçļĦ
    0.26
     it
    0.23
    å®ĥ们
    0.23
    åħ¶
    0.22
     nó
    0.20
     its
    0.19
    ä½Ĩå®ĥ
    0.18
    Activations Density 0.553%

    No Known Activations