© 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. 30779
    Prev
    Next
    INDEX
    Explanations

    Charles Dickens and Guardian

    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
    踵
    -0.11
    éĶ¦
    -0.10
    éĵº
    -0.10
     вÑģÑİ
    -0.09
     зай
    -0.09
    ä¿Ŀä¿®
    -0.09
    ëĿł
    -0.09
    ACES
    -0.09
    unas
    -0.08
    åģ¿
    -0.08
    POSITIVE LOGITS
     Tight
    0.09
    éĩijæ²Ļ
    0.09
     Iterable
    0.08
    贪婪
    0.08
    _HEX
    0.08
    å·´å·´
    0.08
     Portable
    0.08
    zo
    0.08
     Dice
    0.08
     DeV
    0.08
    Activations Density 0.007%

    No Known Activations