© 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. 49288
    Prev
    Next
    INDEX
    Explanations

    wood construction and furniture

    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
    å¾Ģ
    -0.09
    å¸ĥ
    -0.09
    æĶ¶è´¹ç«Ļ
    -0.08
    ounds
    -0.08
    åĨ·åĨ·
    -0.07
    ç®ĵ
    -0.07
    ç»ĿæľĽ
    -0.07
    æѦåĬŁ
    -0.07
    æѦè£ħ
    -0.07
    styl
    -0.07
    POSITIVE LOGITS
    æľ¨è´¨
    0.18
     деÑĢев
    0.17
     wooden
    0.13
    æľ¨
    0.13
    æľ¨æĿIJ
    0.13
     gá»Ĺ
    0.13
    æŀĭ
    0.12
    â½Ĭ
    0.12
    楸
    0.12
     Wooden
    0.11
    Activations Density 0.160%

    No Known Activations