© 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. 38794
    Prev
    Next
    INDEX
    Explanations

    first-person self-reference

    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
    BOR
    -0.11
    è¿ĻåĿĹ
    -0.11
    è¿Ļ件
    -0.10
    æ³Ľ
    -0.09
    lac
    -0.09
    åĿĿ
    -0.09
    åı£æ°Ķ
    -0.09
    桦
    -0.09
    åıĪèĥ½
    -0.09
    åĭ¿
    -0.08
    POSITIVE LOGITS
    æĪijèĩªå·±
    0.12
    ä¹Łæĺ¯å¦ĤæѤ
    0.12
     myself
    0.12
    ä¹Łä¸įä¾ĭå¤ĸ
    0.11
     personally
    0.11
    ç§ģãĤĤ
    0.10
    successful
    0.10
    æ¡Ī
    0.10
    ä¾ĭå¦Ĥ
    0.09
    è¿Ļç§įæĥħåĨµ
    0.09
    Activations Density 0.118%

    No Known Activations