© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50756
    Prev
    Next
    INDEX
    Explanations

    colors: red, orange, green, rosa

    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
    ColumnInfo
    -0.10
    ÑĨион
    -0.09
    é¹ij
    -0.09
    ieres
    -0.09
    chin
    -0.09
    .vue
    -0.09
    åijľ
    -0.09
    TriState
    -0.09
    æĸ¼æĺ¯
    -0.09
    æŁIJ个
    -0.09
    POSITIVE LOGITS
    èĵĿèī²
    0.11
    代表
    0.10
    表示
    0.09
    çĹħä¾ĭ
    0.09
    åħħå®ŀ
    0.09
    æłĩè®°
    0.09
     Jens
    0.09
    æĦŁåı¹
    0.09
     highlighted
    0.08
     represent
    0.08
    Activations Density 0.093%

    No Known Activations