© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6917
    Prev
    Next
    INDEX
    Explanations

    wrong or right

    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
    uzzi
    -0.09
    ularity
    -0.09
    cape
    -0.09
    强è¡Į
    -0.09
    -ST
    -0.09
    æķ´ä¸ª
    -0.09
    ÑįÑĤ
    -0.09
     ת×ŀ
    -0.08
    åij¨åĽ´
    -0.08
    大ä¼Ĺ
    -0.08
    POSITIVE LOGITS
    åIJĪéĢĤçļĦ
    0.28
    æŃ£ç¡®çļĦ
    0.27
    æģ°å½ĵ
    0.27
    æŃ£ç¡®
    0.25
     correct
    0.24
    åIJĪéĢĤ
    0.24
     appropriate
    0.23
     proper
    0.21
    æŃ£ãģĹãģĦ
    0.21
    éĢĤå½ĵçļĦ
    0.21
    Activations Density 0.097%

    No Known Activations