© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8401
    Prev
    Next
    INDEX
    Explanations

    dust, dirt, debris removal

    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
    绽
    -0.10
    äºĨæĪijçļĦ
    -0.09
    cis
    -0.09
    çį¾
    -0.08
    åīģ
    -0.08
    éĢĢåĽŀ
    -0.08
     Buzz
    -0.08
    æ»´æ»´
    -0.08
    hea
    -0.08
     Bail
    -0.08
    POSITIVE LOGITS
    çģ°å°ĺ
    0.26
     dust
    0.25
    å°ĺ
    0.22
     Dust
    0.20
    ç²īå°ĺ
    0.20
    塵
    0.19
    éĻ¤å°ĺ
    0.17
     bụi
    0.16
     dirt
    0.16
    èĦı
    0.15
    Activations Density 0.225%

    No Known Activations