© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 699
    Prev
    Next
    INDEX
    Explanations

    daydreaming, distracted, thinking, lost in thoughts

    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
    Negative Logits
    æŁ¿
    -0.09
    æī®
    -0.09
    æħij
    -0.08
     Hawks
    -0.08
    ản
    -0.08
    æī¬
    -0.08
    åĬ©
    -0.08
    ptr
    -0.08
    æĭį
    -0.08
    zo
    -0.08
    POSITIVE LOGITS
    æĢĿ绪
    0.19
    éģIJ
    0.15
    æĢĿèĢĥ
    0.14
    æĢĿç´¢
    0.13
     thoughts
    0.12
    å¿ĥæĢĿ
    0.11
    æĢĿæĥ³
    0.11
    åĽŀå¿Ĩ
    0.11
    æ²īæµ¸åľ¨
    0.11
    èĦijæµ·ä¸Ń
    0.11
    Activations Density 0.040%

    No Known Activations