© 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. 9807
    Prev
    Next
    INDEX
    Explanations

    aroma and aromatic

    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
    obb
    -0.09
    wise
    -0.09
    æ¨Ĺ
    -0.09
    Å¡ÃŃ
    -0.09
    outu
    -0.09
    utor
    -0.09
    对äºİæĪij们
    -0.09
    ëĭ·
    -0.09
    rames
    -0.08
    idot
    -0.08
    POSITIVE LOGITS
    vore
    0.09
    åħĪæĺ¯
    0.09
    æĢ«
    0.09
    éħ¯
    0.09
    çī¹åĪ«å£°æĺİ
    0.09
     á»ķn
    0.09
    æµ·æĭĶ
    0.09
    çĺ´
    0.08
    éļĻ
    0.08
    ald
    0.08
    Activations Density 0.004%

    No Known Activations