© 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. 59068
    Prev
    Next
    INDEX
    Explanations

    malicious or malignant content/programs

    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
    éĺµ
    -0.11
    åħ¬æĬ¥
    -0.10
    ëĦIJ
    -0.09
    ç½¢
    -0.09
    å¼Ħ
    -0.09
    é¼ĵèĪŀ
    -0.09
    æĸ¯åŁº
    -0.09
    -Clause
    -0.08
    éĻ£
    -0.08
    è¦Ĩ
    -0.08
    POSITIVE LOGITS
    å«Įçĸij
    0.14
    æľī害
    0.12
     suspicious
    0.12
    伪
    0.11
    åı¯çĸij
    0.11
    åį±éĻ©
    0.11
    éĻĮçĶŁ
    0.11
    icious
    0.11
    isky
    0.09
    æ½ľåľ¨
    0.09
    Activations Density 0.111%

    No Known Activations