© 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. 40365
    Prev
    Next
    INDEX
    Explanations

    grow and expand

    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
    èĥľè´Ł
    -0.09
    æķıæĦŁ
    -0.09
    æľ¬ç½ij
    -0.09
    æĤŁ
    -0.09
    latent
    -0.08
    éģģ
    -0.08
    çİĭåŃIJ
    -0.08
    ç£ĭ
    -0.08
    è¿Ļ个èģĮä¸ļ
    -0.08
    allocated
    -0.08
    POSITIVE LOGITS
    æī©å±ķ
    0.16
    æī©å¼ł
    0.16
    æĵ´
    0.14
     expansion
    0.13
     expand
    0.13
    æĭĵå±ķ
    0.13
     Expansion
    0.12
     expands
    0.12
     expanding
    0.12
    åıijå±ķ壮大
    0.11
    Activations Density 0.085%

    No Known Activations