© 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. 33528
    Prev
    Next
    INDEX
    Explanations

    trojan horse

    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
    ya
    -0.10
    ļĮ
    -0.09
     Reyn
    -0.08
    arus
    -0.08
    Ñĭй
    -0.08
    æ¹¾åĮº
    -0.08
    ALLY
    -0.08
    èī¯
    -0.08
    ayar
    -0.08
    yon
    -0.08
    POSITIVE LOGITS
    ä¾¿å®ľ
    0.10
    æĪ·
    0.10
    ieten
    0.09
    éĢģåĩº
    0.09
    iked
    0.09
    èĶ¬
    0.09
    angel
    0.08
     pap
    0.08
    nock
    0.08
    天使
    0.08
    Activations Density 0.003%

    No Known Activations