© 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. 9994
    Prev
    Next
    INDEX
    Explanations

    news service or agency

    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
     hã
    -0.09
     Leone
    -0.09
     اÙĦÙħر
    -0.09
    éªĭ
    -0.09
     ÑģобÑģÑĤвенно
    -0.09
    ä¹į
    -0.09
    æŀ£
    -0.08
    xCB
    -0.08
    edException
    -0.08
    leans
    -0.08
    POSITIVE LOGITS
    èľī
    0.09
    讨
    0.08
    izo
    0.08
    第ä¸Ģ天
    0.08
    Hier
    0.08
    åı·
    0.08
     intrusive
    0.08
    è½Ń
    0.08
    itan
    0.08
    elly
    0.08
    Activations Density 0.009%

    No Known Activations