metadata
library_name: transformers
license: apache-2.0
base_model: HuggingFaceTB/SmolLM2-360M-Instruct
tags:
- generated_from_trainer
model-index:
- name: SlimPajama_10k_SmolLM2
results: []
SlimPajama_10k_SmolLM2
This model is a fine-tuned version of HuggingFaceTB/SmolLM2-360M-Instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 6.1564
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: 5
- eval_batch_size: 5
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.797 | 0.1 | 200 | 7.2221 |
| 7.027 | 0.2 | 400 | 6.8442 |
| 6.8007 | 0.3 | 600 | 6.6666 |
| 6.5956 | 0.4 | 800 | 6.5034 |
| 6.4728 | 0.5 | 1000 | 6.3831 |
| 6.3834 | 0.6 | 1200 | 6.2946 |
| 6.2918 | 0.7 | 1400 | 6.2285 |
| 6.2077 | 0.8 | 1600 | 6.1818 |
| 6.1528 | 0.9 | 1800 | 6.1604 |
| 6.1992 | 1.0 | 2000 | 6.1564 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0