© 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. 7064
    Prev
    Next
    INDEX
    Explanations

    picture, photograph, photo

    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
    åIJijåīį
    -0.10
    TER
    -0.09
    ä¾ĭè¡Į
    -0.08
    chio
    -0.08
    .gt
    -0.08
    çļĦä¿¡æģ¯
    -0.08
    åĩĨå¤ĩ好
    -0.08
    æīĢè°ĵçļĦ
    -0.08
    è¿Ļåıª
    -0.08
    éĺ´
    -0.08
    POSITIVE LOGITS
    æı©
    0.09
    亲åŃIJ
    0.09
    aland
    0.09
    .src
    0.09
    åªļ
    0.09
    æľį
    0.08
    仪
    0.08
    èģĶ
    0.08
    ewise
    0.08
    oss
    0.08
    Activations Density 0.018%

    No Known Activations