© 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. 33944
    Prev
    Next
    INDEX
    Explanations

    kids and toys

    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
    éĿĴå°ijå¹´
    -0.10
    icator
    -0.10
     crossword
    -0.09
     leisure
    -0.09
    -New
    -0.09
    oen
    -0.09
    èĭŁ
    -0.09
    è§Ĥåħī
    -0.09
    ä¸Ńå°ıåѦ
    -0.08
    æĬ±
    -0.08
    POSITIVE LOGITS
    å½±
    0.11
    ç¥ŀåĻ¨
    0.09
    åĦĭ
    0.09
    æĩĭ
    0.09
    éĺ´
    0.09
    bak
    0.09
    åĶ§
    0.09
    rais
    0.09
    龸çİĭ
    0.09
    åı£æ°´
    0.08
    Activations Density 0.094%

    No Known Activations