SmolLM2-135M-Instruct-Arabic-test
This model is a fine-tuned version of HuggingFaceTB/SmolLM2-135M-Instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2311
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.PAGED_ADAMW with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.6658 | 0.0889 | 100 | 1.6589 |
| 1.6393 | 0.1778 | 200 | 1.5819 |
| 1.5377 | 0.2667 | 300 | 1.5419 |
| 1.5329 | 0.3556 | 400 | 1.5107 |
| 1.4348 | 0.4444 | 500 | 1.4799 |
| 1.4884 | 0.5333 | 600 | 1.4576 |
| 1.3968 | 0.6222 | 700 | 1.4366 |
| 1.432 | 0.7111 | 800 | 1.4194 |
| 1.4154 | 0.8 | 900 | 1.4033 |
| 1.3925 | 0.8889 | 1000 | 1.3879 |
| 1.4315 | 0.9778 | 1100 | 1.3771 |
| 1.3474 | 1.0667 | 1200 | 1.3649 |
| 1.3419 | 1.1556 | 1300 | 1.3524 |
| 1.358 | 1.2444 | 1400 | 1.3450 |
| 1.3034 | 1.3333 | 1500 | 1.3356 |
| 1.2675 | 1.4222 | 1600 | 1.3258 |
| 1.3282 | 1.5111 | 1700 | 1.3186 |
| 1.2946 | 1.6 | 1800 | 1.3098 |
| 1.2936 | 1.6889 | 1900 | 1.3027 |
| 1.2275 | 1.7778 | 2000 | 1.2956 |
| 1.2438 | 1.8667 | 2100 | 1.2915 |
| 1.2819 | 1.9556 | 2200 | 1.2826 |
| 1.2551 | 2.0444 | 2300 | 1.2767 |
| 1.2067 | 2.1333 | 2400 | 1.2799 |
| 1.2313 | 2.2222 | 2500 | 1.2687 |
| 1.2186 | 2.3111 | 2600 | 1.2633 |
| 1.2393 | 2.4 | 2700 | 1.2587 |
| 1.1746 | 2.4889 | 2800 | 1.2507 |
| 1.2061 | 2.5778 | 2900 | 1.2494 |
| 1.2453 | 2.6667 | 3000 | 1.2434 |
| 1.1822 | 2.7556 | 3100 | 1.2414 |
| 1.1563 | 2.8444 | 3200 | 1.2335 |
| 1.173 | 2.9333 | 3300 | 1.2311 |
Framework versions
- PEFT 0.14.0
- Transformers 4.51.1
- Pytorch 2.5.1+cu124
- Datasets 3.5.0
- Tokenizers 0.21.0