© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 34641
    Prev
    Next
    INDEX
    Explanations

    or

    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
    .friends
    -0.10
    Friends
    -0.10
    Ø®ÙĬÙĦ
    -0.09
     wsz
    -0.09
    friends
    -0.09
    PERT
    -0.09
    åĢı
    -0.08
    æľī害
    -0.08
    æĿ°
    -0.08
     friends
    -0.08
    POSITIVE LOGITS
    æľĢæĸ°çļĦ
    0.10
    åı±
    0.10
    ä¹İ
    0.09
    (es
    0.09
    ä¹Łä¸įæĥ³
    0.08
    è¯ģåĪ¸äº¤æĺĵ
    0.08
    åį³å°ĨåĪ°æĿ¥
    0.08
    oot
    0.08
    éĢ¼
    0.08
    itage
    0.08
    Activations Density 0.131%

    No Known Activations