© 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. 49330
    Prev
    Next
    INDEX
    Explanations

    numerous followed by noun

    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
    ãģ¤ãģĦ
    -0.10
     notion
    -0.09
    ual
    -0.08
    ows
    -0.08
    åįģ足
    -0.08
    igo
    -0.08
    otropic
    -0.08
    ण
    -0.08
    æĢ»çļĦ
    -0.08
    orta
    -0.08
    POSITIVE LOGITS
    íŀĪ
    0.12
     sclerosis
    0.11
    ä½ķ度ãĤĤ
    0.10
    prech
    0.10
    ethod
    0.09
    -choice
    0.09
    大大å°ıå°ı
    0.09
    hic
    0.09
    обоÑĢоÑĤ
    0.09
    ely
    0.09
    Activations Density 0.014%

    No Known Activations