© 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. 1746
    Prev
    Next
    INDEX
    Explanations

    first goal

    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
     Stap
    -0.10
    bos
    -0.10
    éģĩ
    -0.09
    bon
    -0.09
    è¿ĺ好
    -0.09
    Ñħа
    -0.09
    osen
    -0.09
     bon
    -0.08
    åij¨
    -0.08
    æĸ¹ä¾¿
    -0.08
    POSITIVE LOGITS
    大éĥ¨åĪĨ
    0.10
    å·±
    0.10
    ansen
    0.09
    æİIJ
    0.09
     Recon
    0.09
    æ´»
    0.09
    ä¸Ģ大æĹ©
    0.08
    追éĢIJ
    0.08
    ات
    0.08
    ä¼¼
    0.08
    Activations Density 0.021%

    No Known Activations