© 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. 59279
    Prev
    Next
    INDEX
    Explanations

    reduce, smaller, down

    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
    osit
    -0.10
    krit
    -0.10
    ิà¸ķ
    -0.10
    iked
    -0.09
    ossed
    -0.09
    assel
    -0.09
    å¿ĺäºĨ
    -0.09
    汩
    -0.09
    iences
    -0.09
    殿ä¸ĭ
    -0.09
    POSITIVE LOGITS
    åĩıå°ijäºĨ
    0.17
    åĩıå°ij
    0.15
    éĻįä½İ
    0.13
     reduced
    0.13
    åīĬåĩı
    0.13
    æ¸Ľå°ij
    0.13
    缩åĩı
    0.12
     reduce
    0.12
     redu
    0.12
     reduces
    0.12
    Activations Density 0.197%

    No Known Activations