© 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. 21570
    Prev
    Next
    INDEX
    Explanations

    laser skin

    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
    éĢĭ
    -0.10
    æıIJçĤ¼
    -0.09
    åĢº
    -0.09
    æīİ
    -0.08
     unt
    -0.08
    .mk
    -0.08
     Toledo
    -0.08
     Official
    -0.08
    ä¸ĩäºĭ
    -0.08
    æ¼Ĩ
    -0.08
    POSITIVE LOGITS
    åĪ¹
    0.12
    ä¼IJ
    0.11
    ØŃض
    0.10
    åıĪè¦ģ
    0.09
    oglob
    0.09
    åºĶ对
    0.09
    åij²
    0.09
    派人
    0.08
    rush
    0.08
    çİĦ
    0.08
    Activations Density 0.022%

    No Known Activations