Update ML Intern artifact metadata
Browse files
README.md
CHANGED
|
@@ -1,3 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# ModernProteinLM: Next-Generation Protein Encoder
|
| 2 |
|
| 3 |
A next-generation protein language model architecture that combines state-of-the-art NLP encoder improvements with protein-specific training innovations to push predictive task performance under 200M parameters.
|
|
@@ -169,3 +173,23 @@ If you use this architecture, cite:
|
|
| 169 |
- ELECTRA (Clark et al., ICLR 2020)
|
| 170 |
- NeoBERT (2025)
|
| 171 |
- SpanBERT (Joshi et al., 2020)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- ml-intern
|
| 4 |
+
---
|
| 5 |
# ModernProteinLM: Next-Generation Protein Encoder
|
| 6 |
|
| 7 |
A next-generation protein language model architecture that combines state-of-the-art NLP encoder improvements with protein-specific training innovations to push predictive task performance under 200M parameters.
|
|
|
|
| 173 |
- ELECTRA (Clark et al., ICLR 2020)
|
| 174 |
- NeoBERT (2025)
|
| 175 |
- SpanBERT (Joshi et al., 2020)
|
| 176 |
+
|
| 177 |
+
<!-- ml-intern-provenance -->
|
| 178 |
+
## Generated by ML Intern
|
| 179 |
+
|
| 180 |
+
This model repository was generated by [ML Intern](https://github.com/huggingface/ml-intern), an agent for machine learning research and development on the Hugging Face Hub.
|
| 181 |
+
|
| 182 |
+
- Try ML Intern: https://smolagents-ml-intern.hf.space
|
| 183 |
+
- Source code: https://github.com/huggingface/ml-intern
|
| 184 |
+
|
| 185 |
+
## Usage
|
| 186 |
+
|
| 187 |
+
```python
|
| 188 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 189 |
+
|
| 190 |
+
model_id = 'GrimSqueaker/ModernProteinLM'
|
| 191 |
+
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
| 192 |
+
model = AutoModelForCausalLM.from_pretrained(model_id)
|
| 193 |
+
```
|
| 194 |
+
|
| 195 |
+
For non-causal architectures, replace `AutoModelForCausalLM` with the appropriate `AutoModel` class.
|