© 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. 32457
    Prev
    Next
    INDEX
    Explanations

    actually + verb

    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
    è½»æĺĵ
    -0.10
    çĶļèĩ³æĺ¯
    -0.10
     tháºŃm
    -0.10
    çĶļèĩ³
    -0.09
    ancia
    -0.09
     purely
    -0.09
    thon
    -0.09
     ÑģобÑģÑĤвенно
    -0.09
    ä¹ĭäºī
    -0.09
    ãģ¨ãģ«ãģĭãģı
    -0.09
    POSITIVE LOGITS
    -minded
    0.11
    ity
    0.10
    æĦıä¹īä¸ĬçļĦ
    0.09
     Ritch
    0.09
    ä½ľèĢħæľ¬äºº
    0.09
    ç½ķ
    0.09
    Temporal
    0.09
    ele
    0.09
    ially
    0.09
    æĺ¯æľī
    0.08
    Activations Density 0.013%

    No Known Activations