© 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. 63627
    Prev
    Next
    INDEX
    Explanations

    images in empty boxes

    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
    åľ¨éĩĮéĿ¢
    -0.10
    æĬĦ
    -0.10
     Blind
    -0.09
    ivor
    -0.09
    çĭ·
    -0.09
    .getMethod
    -0.09
    ROID
    -0.09
    áŃ
    -0.08
     Fountain
    -0.08
    æĺ¯ä¸Ģ款
    -0.08
    POSITIVE LOGITS
    åİŁå§ĭ
    0.11
    -image
    0.09
     image
    0.09
    atts
    0.09
    .Image
    0.09
    å¤ļåªĴä½ĵ
    0.09
     setImage
    0.09
    è§Ĩè§ī
    0.08
    ãĤ¤ãĥ¡ãĥ¼ãĤ¸
    0.08
     Image
    0.08
    Activations Density 0.442%

    No Known Activations