© 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. 18394
    Prev
    Next
    INDEX
    Explanations

    save the world or items

    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¸įå®ī
    -0.10
    å¿Į
    -0.08
     ÑģовеÑĢÑĪен
    -0.08
    esor
    -0.08
     ÑįкÑģпеÑĢÑĤ
    -0.08
    å®īåħ¨ä¿Ŀéļľ
    -0.08
    sis
    -0.08
    duino
    -0.08
    èµ·æºIJ
    -0.08
    enna
    -0.08
    POSITIVE LOGITS
    æĢ§åij½
    0.13
    (save
    0.10
    æĭ¯æķij
    0.10
     lives
    0.10
    æĬ¢æķij
    0.10
     Lives
    0.10
    èµİ
    0.09
    æĮ½æķij
    0.09
     saves
    0.09
    .Save
    0.09
    Activations Density 0.095%

    No Known Activations