© 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. 32833
    Prev
    Next
    INDEX
    Explanations

    non-English characters

    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
    3
    -0.08
    2
    -0.08
    5
    -0.08
     re
    -0.07
    4
    -0.07
     Fence
    -0.07
    9
    -0.07
     nu
    -0.07
    7
    -0.07
     Tops
    -0.07
    POSITIVE LOGITS
    佬
    0.11
    ç»ıå¼Ģ
    0.11
    躯
    0.10
     jose
    0.10
    巨头
    0.10
    è¿ĻäºĽä¸ľè¥¿
    0.10
    æ´²
    0.10
     cái
    0.10
    è¯ij
    0.10
    èĭ±çī¹
    0.09
    Activations Density 0.257%

    No Known Activations