zephyr-7b-math-case-6
This model is a fine-tuned version of alignment-handbook/zephyr-7b-sft-full on the EunsuKim/GSM8K and the EunsuKim/MATH datasets.
It achieves the following results on the evaluation set:
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 64
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 10
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
| 0.8827 |
1.0 |
5 |
0.7429 |
| 0.6709 |
2.0 |
10 |
0.5531 |
| 0.5071 |
3.0 |
15 |
0.4035 |
| 0.3519 |
4.0 |
20 |
0.2455 |
| 0.2036 |
5.0 |
25 |
0.1302 |
| 0.1035 |
6.0 |
30 |
0.0602 |
| 0.0527 |
7.0 |
35 |
0.0356 |
| 0.0321 |
8.0 |
40 |
0.0249 |
| 0.0236 |
9.0 |
45 |
0.0206 |
| 0.0202 |
10.0 |
50 |
0.0198 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1
