© 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. 33628
    Prev
    Next
    INDEX
    Explanations

    questions about sentences

    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
    éĵŃ
    -0.08
    ãģ®ãģł
    -0.08
    æķ¬è¯·
    -0.08
    æıĸ
    -0.08
     flex
    -0.08
    rror
    -0.07
    ç»ĥ
    -0.07
    1
    -0.07
    æIJĶ
    -0.07
     ÑģÑĦ
    -0.07
    POSITIVE LOGITS
    è¿Ļåı¥è¯Ŀ
    0.12
    çļĦæĦıæĢĿ
    0.10
    ä»Ģä¹ĪæĦıæĢĿ
    0.10
    æĮĩçļĦæĺ¯
    0.10
    éĻĪè¿°
    0.09
    éĿĽ
    0.09
    BITS
    0.09
    æĦıæĢĿ
    0.09
    HorizontalAlignment
    0.09
    ores
    0.09
    Activations Density 0.329%

    No Known Activations