© 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. 24239
    Prev
    Next
    INDEX
    Explanations

    "yoda" or "schadenfreude"

    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
    åĺ¤
    -0.10
     outside
    -0.09
    èł¡
    -0.09
     Outside
    -0.09
    ongan
    -0.08
    Hur
    -0.08
    æIJ¬å®¶
    -0.08
     Hur
    -0.08
    dash
    -0.08
    åį°
    -0.08
    POSITIVE LOGITS
    黳
    0.10
    å¹½é»ĺ
    0.10
    rig
    0.10
    çªĺ
    0.10
    eks
    0.10
    zilla
    0.09
    æIJŀç¬ij
    0.09
    ä¼ģ
    0.09
    æĮł
    0.09
    .Setup
    0.09
    Activations Density 0.072%

    No Known Activations