© 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. 1887
    Prev
    Next
    INDEX
    Explanations

    starts off

    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
    æ¶Ī
    -0.10
    ç½ijé¦ĸ页
    -0.10
    اÙħت
    -0.10
    æ¡Ĩ
    -0.09
    .priv
    -0.09
    涮
    -0.09
    ınt
    -0.09
    åĪļ好
    -0.09
    æŸ
    -0.09
    å¼Ģåı£
    -0.08
    POSITIVE LOGITS
    ripp
    0.11
     promise
    0.09
     promising
    0.09
    åħ¸åŀĭçļĦ
    0.09
     dem
    0.09
    åIJ¬èµ·æĿ¥
    0.09
    ç¨ļ
    0.08
    佶
    0.08
    å¾Īç®Ģåįķ
    0.08
    å´Ľ
    0.08
    Activations Density 0.030%

    No Known Activations