© 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. 27323
    Prev
    Next
    INDEX
    Explanations

    alignment issues

    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
    references
    -0.10
    okers
    -0.09
    è¿Ĥ
    -0.09
    å¸Ńåį·
    -0.09
    å°ıèħ¿
    -0.09
    弥漫
    -0.09
    ä¸į容
    -0.09
    ç¦ģåĮº
    -0.09
    æ¯ı人
    -0.09
    oku
    -0.09
    POSITIVE LOGITS
    æŃª
    0.16
    èĦ±èIJ½
    0.13
    æĿ¾
    0.10
    æİīèIJ½
    0.10
    ignon
    0.10
    _ALIGNMENT
    0.10
    çī¢åĽº
    0.09
    åĢ¾æĸľ
    0.09
     alignment
    0.09
    astle
    0.08
    Activations Density 0.208%

    No Known Activations