© 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. 1377
    Prev
    Next
    INDEX
    Explanations

    stabbing knives chased fuel flashed

    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
    -0.10
    åĴĮè°IJ
    -0.09
    ru
    -0.09
     ден
    -0.09
     Isles
    -0.09
    ned
    -0.09
     Cheer
    -0.08
    spar
    -0.08
    inspace
    -0.08
    zept
    -0.08
    POSITIVE LOGITS
    æݬ
    0.10
    侮辱
    0.10
    æĿ¿åĿĹ
    0.09
    çıī
    0.09
    æijĶ
    0.09
    oard
    0.09
    amble
    0.09
    æIJŃè½½
    0.09
    主èIJ¥
    0.08
    è¿©
    0.08
    Activations Density 0.098%

    No Known Activations