```markdown SKEP Model Summary ————————————
The following table summarizes the currently supported SKEP models and their corresponding pretrained weights in PaddleNLP. For detailed model specifications, please refer to the corresponding links.
Notes:
1. All code blocks (e.g., skep_ernie_1.0_large_ch) and technical terms (e.g., “PaddleNLP”) are preserved in original English.
2. Markdown formatting (tables, code blocks, headers) is maintained exactly as in the source.
3. Links and anchors (e.g., [link](#here)) remain unchanged per requirements.
4. Academic grammar and technical writing style are applied throughout the translation.
skep_roberta_large_en | English | 24-layer, 1024-hidden, |
| | | 16-heads, 336M parameters. |
| | | Trained using the RoBERTa model |
| | | roberta_large |
+———————————————————————————-+————–+———————————————————————————-+
| English | | |
+———————————————————————————-+————–+———————————————————————————-+
| 24-layer, 1024-hidden, | | 16-heads, 355M parameters |
| Trained using the RoBERTa model | | roberta_large_en |
+———————————————————————————-+————–+———————————————————————————-+