© 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. 3945
    Prev
    Next
    INDEX
    Explanations

    years or periods

    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
    anka
    -0.10
    产çĶŁçļĦ
    -0.09
    velt
    -0.09
    otta
    -0.09
    æ¯Ķè¾ĥé«ĺ
    -0.08
    产åIJİ
    -0.08
    åIJĪä½ľ
    -0.08
    å½Ĵ
    -0.08
     Petit
    -0.08
    amen
    -0.08
    POSITIVE LOGITS
    æijģ
    0.09
    æĴ¼
    0.09
    好åĩłä¸ª
    0.09
    åij¨è½¬
    0.09
     Peters
    0.08
    款
    0.08
    ä¸Ģ段æŶéĹ´
    0.08
    åĪ¶
    0.08
    两个æľĪ
    0.08
    ngr
    0.08
    Activations Density 0.031%

    No Known Activations