© 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. 29639
    Prev
    Next
    INDEX
    Explanations

    measure, evaluate, results, effectiveness

    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
    æĮŁ
    -0.09
    大åħ¨
    -0.09
     ден
    -0.09
    åĿĩåĮĢ
    -0.08
    RuleContext
    -0.08
    ä¸ĭè°ĥ
    -0.08
    æĪı
    -0.08
     niche
    -0.08
    å¤Ļ
    -0.08
    窥
    -0.08
    POSITIVE LOGITS
    æĪIJæķĪ
    0.17
    æķĪæŀľ
    0.16
    å®ŀæķĪ
    0.16
    æĪIJæŀľ
    0.12
     outcomes
    0.12
    ç»ĵæŀľ
    0.12
    è¡¡éĩı
    0.12
     effectiveness
    0.11
     outcome
    0.11
     metrics
    0.11
    Activations Density 0.052%

    No Known Activations