© 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. 40034
    Prev
    Next
    INDEX
    Explanations

    speaker or speakers

    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
     vit
    -0.09
    èĨº
    -0.08
    åħīçĶµ
    -0.08
    夸
    -0.08
     lam
    -0.08
    atem
    -0.08
    wap
    -0.08
    easy
    -0.08
    æ°´èµĦæºIJ
    -0.08
    åģļå®Į
    -0.08
    POSITIVE LOGITS
    èī¯
    0.11
    声éģĵ
    0.10
    ä¸Ģéĺµ
    0.09
    éŁ³åĵį
    0.09
    èĪ·
    0.09
     Malone
    0.09
    山路
    0.09
     Chapman
    0.09
    systems
    0.08
     Blow
    0.08
    Activations Density 0.071%

    No Known Activations