© 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. 29846
    Prev
    Next
    INDEX
    Explanations

    smell or odor

    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
    èī²å½©
    -0.10
    ayment
    -0.10
     Sabb
    -0.09
    MBER
    -0.09
    yre
    -0.09
     há»ı
    -0.09
     Saul
    -0.09
    å³ĭ
    -0.08
    äºŀ
    -0.08
    FUN
    -0.08
    POSITIVE LOGITS
    iest
    0.10
    åķĨåĵģæĪ¿
    0.10
    ened
    0.10
    æĭ±
    0.10
    ening
    0.09
    åĩŃè¯ģ
    0.09
    lessly
    0.09
    less
    0.09
    æľĪä¸Ń
    0.08
    è§ļ
    0.08
    Activations Density 0.029%

    No Known Activations