Update README.md

README.md

@@ -28,7 +28,7 @@ Sequences were:
 
 ### Loading the Model
 
-\`\`\`python
+```python
 from transformers import EsmModel, AutoTokenizer
 import torch
 
@@ -36,11 +36,11 @@ import torch
 model = EsmModel.from_pretrained("MahTala/AbCDR-ESM2")
 tokenizer = AutoTokenizer.from_pretrained("MahTala/AbCDR-ESM2")
 model.eval()
-\`\`\`
+```
 
 ### Extract Embeddings
 
-\`\`\`python
+```python
 # Prepare paired sequence
 SEP_TOKEN = "-" 
 heavy_chain = (
@@ -66,7 +66,7 @@ mask = inputs["attention_mask"].unsqueeze(-1)
 pooled = (embeddings * mask).sum(1) / mask.sum(1)
 
 print(f"Embedding shape: {pooled.shape}")  # (1, 2560)
-\`\`\`
+```
 
 ## Input Format
 
@@ -78,9 +78,9 @@ print(f"Embedding shape: {pooled.shape}")  # (1, 2560)
 - Uncommon residues should be replaced with X
 
 **Example:**
-\`\`\`python
+```python
 sequence = "EVQLVESGGGLVQPGGSLRLSCAASGFTFSSYAMS...-DIQMTQSPSSLSASVGDRVTITCRASQSISS..."
-\`\`\`
+```
 
 ## Output