Upload README.md with huggingface_hub

README.md

@@ -7,18 +7,18 @@ tags: []
 The GitHub with the implementation and requirements.txt can be found [here](https://github.com/Synthyra/FastPLMs.git)
 
 # Profluent-E1
-[Profluent-E1](https://github.com/Synthyra/ESMplusplus) is a faithful implementation of Profluent's models [E1](https://www.profluent.bio/showcase/e1) ([license](https://github.com/Profluent-AI/E1/tree/main?tab=License-1-ov-file)) that integrates Huggingface AutoModel compatability and nice embedding functionality built in.
+[Profluent-E1](https://github.com/Synthyra/ESMplusplus) is a faithful implementation of Profluent's [E1](https://www.profluent.bio/showcase/e1) models ([license](https://github.com/Profluent-AI/E1/tree/main?tab=License-1-ov-file)) that integrates Huggingface AutoModel compatability and nice embedding functionality.
 
 
 ## Use with 🤗 transformers
 ### Supported models
 ```python
 model_dict = {
-    # Synthyra/Profluent-E1_150M
+    # Synthyra/Profluent-E1-150M
     'Profluent-E1-150M': 'Profluent-Bio/E1-150m',
-    # Synthyra/Profluent-E1_150M
+    # Synthyra/Profluent-E1-150M
     'Profluent-E1-300M': 'Profluent-Bio/E1-300m',
-    # Synthyra/Profluent-E1_150M
+    # Synthyra/Profluent-E1-150M
     'Profluent-E1-600M': 'Profluent-Bio/E1-600m',
 }
 ```
@@ -127,17 +127,6 @@ for param in model.classifier.parameters():
 For a more thourough example of fine-tuning, check out our example script [here](https://github.com/Synthyra/FastPLMs/blob/main/fine_tuning_example.py).
 
 
-## Returning attention maps
-Usually F.scaled_dot_product_attention is used for the attention calculations, which is much faster than native PyTorch. However, it cannot return attention maps.
-ESM++ has the option to ```output_attentions```, which will calculate attention manually. This is much slower, so do not use unless you need the attention maps.
-
-```python
-output = model(**tokenized, output_attentions=True)
-att = output.attentions
-len(att) # 30, one for each layer, size (batch_size, num_heads, seq_len, seq_len) each
-```
-
-
 ### Citation
 If you use any of this implementation or work please cite the following DOI and Profluent's paper.
 
@@ -153,7 +142,8 @@ If you use any of this implementation or work please cite the following DOI and
 ```
 
 ```
- @article{Jain_Beazer_Ruffolo_Bhatnagar_Madani_2025, title={E1: Retrieval-Augmented Protein Encoder Models},
+ @article{Jain_Beazer_Ruffolo_Bhatnagar_Madani_2025,
+    title={E1: Retrieval-Augmented Protein Encoder Models},
     url={https://www.biorxiv.org/content/early/2025/11/13/2025.11.12.688125},
     DOI={10.1101/2025.11.12.688125},
     journal={bioRxiv},