Fela 1.6M
Browse files
README.md
CHANGED
|
@@ -18,6 +18,10 @@ PyTorch written protein language model on the hyena operator (1.6M params)
|
|
| 18 |
- Data: Pfam-A (filtered to 20–512 residues, standard alphabet only), ~9.5B tokens
|
| 19 |
- Training: 40k steps, batch 256, bf16, AdamW (wd 0.1), cosine LR 6e-4 → 6e-5
|
| 20 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
## Config
|
| 22 |
|
| 23 |
| Parameter | Value |
|
|
|
|
| 18 |
- Data: Pfam-A (filtered to 20–512 residues, standard alphabet only), ~9.5B tokens
|
| 19 |
- Training: 40k steps, batch 256, bf16, AdamW (wd 0.1), cosine LR 6e-4 → 6e-5
|
| 20 |
|
| 21 |
+
based on the paper: https://www.biorxiv.org/content/10.1101/2024.01.18.576206v1
|
| 22 |
+
|
| 23 |
+
base model (not finetuned)
|
| 24 |
+
|
| 25 |
## Config
|
| 26 |
|
| 27 |
| Parameter | Value |
|