© 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. 32434
    Prev
    Next
    INDEX
    Explanations

    unsavory, insidious, depravity, rot

    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
    è¿·
    -0.10
    æĸ©
    -0.09
    èµ°åĩºåİ»
    -0.09
    çĹ´
    -0.09
    direct
    -0.09
    æľŁåĨħ
    -0.08
    uesta
    -0.08
     pragma
    -0.08
    çĸ²
    -0.08
    ç»ĵæŀĦæĢ§
    -0.08
    POSITIVE LOGITS
    èĤ®
    0.13
    æģ¶å¿ĥ
    0.12
    çĮ¥
    0.12
    çĭ¡
    0.12
    é¾Į
    0.11
    èĦı
    0.11
    ä¸ij
    0.11
    åįij
    0.10
    éĺ´
    0.10
    dirty
    0.10
    Activations Density 0.038%

    No Known Activations