© 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. 22395
    Prev
    Next
    INDEX
    Explanations

    just or simply

    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
    alia
    -0.10
    åĶ¯ä¸Ģ
    -0.09
    è¦ģä¹Ī
    -0.09
    称
    -0.09
     particular
    -0.09
    fest
    -0.08
    rowing
    -0.08
    说çļĦæĺ¯
    -0.08
    ä¸įæĺİçĻ½
    -0.08
    åĭ¾
    -0.08
    POSITIVE LOGITS
    æľ¬èº«å°±
    0.13
    æľ¬èº«å°±æĺ¯
    0.11
    æĻ®éĢļçļĦ
    0.11
    æľ¬èº«çļĦ
    0.11
    åįķ纯çļĦ
    0.10
    å¼Ģåı£
    0.10
     налиÑĩ
    0.10
    æľ¬èº«
    0.10
    mere
    0.10
    å¼ĢåºĹ
    0.09
    Activations Density 0.035%

    No Known Activations