© 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. 10429
    Prev
    Next
    INDEX
    Explanations

    jokes and humor

    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
    Soft
    -0.09
     soft
    -0.09
    iew
    -0.09
    áĥ¥
    -0.09
    hti
    -0.08
     Soft
    -0.08
    iete
    -0.08
    èĢĹ
    -0.08
     pseud
    -0.08
    è®®äºĭ
    -0.08
    POSITIVE LOGITS
    éĢĤæŶ
    0.12
    åºķçĽĺ
    0.11
    涡
    0.10
    estre
    0.10
    ä¸Ģç¬ij
    0.10
    çİ©ç¬ij
    0.10
    æĭĬ
    0.10
    é¹ij
    0.09
    éĻª
    0.09
    ENSE
    0.09
    Activations Density 0.039%

    No Known Activations