© 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. Qwen3-14B
    5. 20-RESID-BATCHTOPK-65K__L0-80
    7. 16614
    Prev
    Next
    INDEX
    Explanations

    counting, world, way, away, Again, gal, rules, Everyone

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-14b-saes/saes_Qwen_Qwen3-14B_batch_top_k/resid_post_layer_20
    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
     ÐIJлекÑģан
    -0.32
     ÐŁÐ¾Ñĩем
    -0.29
     ÐĶмиÑĤÑĢ
    -0.26
     заÑĢегиÑģÑĤ
    -0.25
     пÑĥÑĤеÑĪ
    -0.23
     Flyers
    -0.22
     пÑĢиÑıÑĤн
    -0.22
     ÐĶм
    -0.21
     меÑĢоп
    -0.21
     PaÅĦstwo
    -0.21
    POSITIVE LOGITS
    æĶ»åĬ¿
    0.21
    æ®Ĥ
    0.19
     неÑģк
    0.18
    !
    0.18
    xae
    0.18
    ä¼łæĿ¥
    0.17
    çıĬ
    0.17
     бÑİджеÑĤ
    0.17
    ç»Ŀä¸į
    0.17
    梦è§ģ
    0.17
    Activations Density 0.079%

    No Known Activations