© 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. 22609
    Prev
    Next
    INDEX
    Explanations

    create one, result ideal, give you

    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
    rint
    -0.09
     Nets
    -0.08
    æľ´
    -0.08
    æ½¼
    -0.08
    andal
    -0.08
    å®ļçĤ¹
    -0.08
    大海
    -0.08
    acket
    -0.08
    éli
    -0.08
     luck
    -0.08
    POSITIVE LOGITS
    åıªèĥ½è¯´
    0.12
    综åIJĪ
    0.11
    èŀįåªĴä½ĵ
    0.10
    综åIJĪä½ĵ
    0.10
    æĹłä¸į
    0.10
    综åIJĪå®ŀåĬĽ
    0.10
    æŀĦæĪIJäºĨ
    0.09
    ä¸ĭæĿ¥çļĦ
    0.09
    å½¢æĪIJäºĨ
    0.09
    tro
    0.09
    Activations Density 0.241%

    No Known Activations