© 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. 57397
    Prev
    Next
    INDEX
    Explanations

    tread carefully or perilously

    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
    èijĹåIJįçļĦ
    -0.10
    ëĿ¼ëıĦ
    -0.10
    ATS
    -0.09
    PEG
    -0.08
    ÂĢÂ
    -0.08
    ity
    -0.08
    çĭłæĬĵ
    -0.08
    éĶ·
    -0.08
    éĶ¢
    -0.08
    ä¿ļ
    -0.08
    POSITIVE LOGITS
    è¸ı
    0.10
    iness
    0.10
    éģĵçIJĨ
    0.10
    è¸ıå®ŀ
    0.10
    uting
    0.09
    æĸ¹å¼ı
    0.09
    ÑĥÑī
    0.09
    æ±Łæ¹ĸ
    0.09
    stakes
    0.09
    è¸ī
    0.09
    Activations Density 0.006%

    No Known Activations