Model Card
- Source: https://arxiv.org/abs/2509.02046
- Optimizer:
mini - Model size:
300m - Data size:
48B
Best configuration
| Hyperparameter | Value |
|---|---|
| beta1 | 0.9 |
| beta2 | 0.98 |
| epsilon | 9.999999999999999e-26 |
| learning_rate | 0.002 |
| max_grad_norm | 2 |
| min_lr_ratio | 0 |
| nesterov | False |
| train_batch_size | 128 |
| warmup | 2000 |
| weight_decay | 0.2 |