© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 47833
    Prev
    Next
    INDEX
    Explanations

    out, out on, out of

    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
    éĢ¼
    -0.09
    ×ķק
    -0.09
    æŀIJ
    -0.08
    人éĥ½
    -0.08
     suite
    -0.08
    æĸĩç«łæĿ¥æºIJ
    -0.08
     Nancy
    -0.08
    æīįç®Ĺ
    -0.08
    çĿ¨
    -0.08
     Suite
    -0.08
    POSITIVE LOGITS
    åºĶæľīçļĦ
    0.10
    orative
    0.09
    ernal
    0.09
    侪
    0.09
    æ¯ĶåĪĨ
    0.08
    失èIJ½
    0.08
    rlen
    0.08
    ively
    0.08
    æīī
    0.08
    appen
    0.08
    Activations Density 0.002%

    No Known Activations