© 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. 50296
    Prev
    Next
    INDEX
    Explanations

    judges

    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
    ision
    -0.10
    hone
    -0.09
     Reader
    -0.09
    ities
    -0.09
    å½¼
    -0.08
    íij¸
    -0.08
     Sitting
    -0.08
    履约
    -0.08
     Franc
    -0.08
     scoop
    -0.08
    POSITIVE LOGITS
    夯
    0.10
    imon
    0.08
    itech
    0.08
    纶
    0.08
     Sham
    0.08
    å§Ķå®£ä¼łéĥ¨
    0.08
    [email
    0.08
    说æĪij
    0.08
    éĿĴçĿIJ
    0.08
    æĪIJåijĺ
    0.08
    Activations Density 0.008%

    No Known Activations