© 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. 32954
    Prev
    Next
    INDEX
    Explanations

    vocal, choir, conductor, gospel

    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
    ç®Ģ便
    -0.09
    éģ¿
    -0.08
    赶
    -0.08
    bsp
    -0.08
    æĸ¹ä¾¿
    -0.08
    ze
    -0.08
    åį±
    -0.08
    ivers
    -0.08
    _PKG
    -0.08
    .isNull
    -0.08
    POSITIVE LOGITS
    æªĢ
    0.11
     hôn
    0.11
    OME
    0.10
    _gid
    0.10
    åı£ç½©
    0.10
    åºŀ大
    0.10
    çĮ®è¡Ģ
    0.09
    å°¸
    0.09
     Harmony
    0.09
    ä¸Ī
    0.09
    Activations Density 0.093%

    No Known Activations