© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7760
    Prev
    Next
    INDEX
    Explanations

    type of

    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
    æĭ¼éŁ³
    -0.09
     overs
    -0.08
    æĹ©æĹ¥
    -0.08
    åķĸ
    -0.08
    çł£
    -0.08
    acen
    -0.08
    abit
    -0.07
     wake
    -0.07
    ç±´
    -0.07
    责任人
    -0.07
    POSITIVE LOGITS
    ç±»åŀĭçļĦ
    0.13
    æ·ij
    0.11
    ãĤ¦ãĥ³
    0.09
    /licenses
    0.09
    æļ´éľ²
    0.09
    èħĶ
    0.09
    à¸Īา
    0.09
     Loc
    0.09
    å¼ĵ
    0.09
    æģģ
    0.09
    Activations Density 0.075%

    No Known Activations