© 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. 22486
    Prev
    Next
    INDEX
    Explanations

    shenanigans, mischief, exploits, antics

    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
    å¥ĹæĪ¿
    -0.10
    è½Ń
    -0.10
    æīĵ好
    -0.10
    è°Īè¯Ŀ
    -0.10
    人éĥ½
    -0.10
    游åĩ»éĺŁ
    -0.10
    è¿Ļæł·æīįèĥ½
    -0.10
    æĹħé¦Ĩ
    -0.09
    è¿Ļ个人
    -0.09
    ylvania
    -0.09
    POSITIVE LOGITS
     mischief
    0.11
    ous
    0.09
     antics
    0.09
    pus
    0.09
    ancel
    0.09
    oba
    0.09
    atches
    0.08
    auled
    0.08
    trimmed
    0.08
    _trim
    0.08
    Activations Density 0.023%

    No Known Activations