© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 29402
    Prev
    Next
    INDEX
    Explanations

    safety harness, rail, blood, lift, device, glass

    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
    Negative Logits
    累累
    -0.11
    åĦ¿
    -0.10
    ä¸įéĢļ
    -0.10
    æł·çļĦ
    -0.10
    TC
    -0.10
    adero
    -0.10
    enden
    -0.09
    inding
    -0.09
    ../../../
    -0.09
    æ¯ĽçĹħ
    -0.09
    POSITIVE LOGITS
    ties
    0.10
    第ä¸Ģ
    0.09
    à¹Į
    0.09
    itas
    0.08
    rog
    0.08
    AGO
    0.08
    æĮ½
    0.08
    åīįèĢħ
    0.08
    ored
    0.08
    ered
    0.08
    Activations Density 0.023%

    No Known Activations