© 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. 18234
    Prev
    Next
    INDEX
    Explanations

    paranoid, suspicious, or fearful states

    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
    éĩĩ
    -0.09
    寮
    -0.09
    æĹłå£°
    -0.09
    ç¥ŀåľ£
    -0.09
    orton
    -0.08
    çļĦ巨大
    -0.08
    æµģåIJij
    -0.08
    oland
    -0.08
    .Abs
    -0.08
    ä¸įå¦Ĥ
    -0.08
    POSITIVE LOGITS
     paranoia
    0.16
     paranoid
    0.15
    çĸij
    0.12
    æķıæĦŁ
    0.12
    çĦ¦èĻij
    0.12
    æĭħå¿ĥ
    0.12
    æĭħ忧
    0.11
    è¿ĩ度
    0.11
    æĢĢçĸij
    0.11
    çĶŁæĢķ
    0.11
    Activations Density 0.044%

    No Known Activations