Update README.md

README.md

@@ -11,6 +11,30 @@ tags: []
 
 ## Model Details
 
+### Code to create models
+```python
+from transformers import ModernBertConfig, ModernBertForSequenceClassification, AutoTokenizer
+
+model_id = "answerdotai/ModernBERT-base"
+config = ModernBertConfig.from_pretrained(
+        model_id,
+        hidden_size=32,
+        intermediate_size=64,
+        num_hidden_layers=4,
+        num_attention_heads=2,
+        global_attn_every_n_layers=3,
+)
+
+# Create model and randomize all weights
+model = ModernBertForSequenceClassification(config)
+
+torch.manual_seed(0) # Set for reproducibility
+for name, param in model.named_parameters():
+    param.data = torch.randn_like(param)
+
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+```
+
 ### Model Description
 
 <!-- Provide a longer summary of what this model is. -->