© 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. 51224
    Prev
    Next
    INDEX
    Explanations

    numbers followed by cases

    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¿¡å¾Ĵ
    -0.10
    æ¼¾
    -0.10
    èħ¿
    -0.09
    /trunk
    -0.09
    éªĿ
    -0.09
    igel
    -0.09
    æĹıèĩªæ²»
    -0.09
    ionales
    -0.09
    æĽ´é«ĺ
    -0.09
    åįı
    -0.09
    POSITIVE LOGITS
    pedia
    0.10
    inde
    0.08
    nox
    0.07
    åij½
    0.07
    abar
    0.07
    hausen
    0.07
     Quincy
    0.07
    alist
    0.07
    /sl
    0.07
     Valle
    0.07
    Activations Density 0.323%

    No Known Activations