mistral-Mistral-Finetune-1
This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.0554
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: 2.5e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1
- num_epochs: 2
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.8941 | 0.05 | 25 | 1.7724 |
1.8315 | 0.1 | 50 | 1.7261 |
1.7522 | 0.14 | 75 | 1.6971 |
1.6974 | 0.19 | 100 | 1.6678 |
1.7149 | 0.24 | 125 | 1.6430 |
1.6037 | 0.29 | 150 | 1.6201 |
1.6611 | 0.34 | 175 | 1.6057 |
1.7131 | 0.38 | 200 | 1.5854 |
1.7619 | 0.43 | 225 | 1.5696 |
1.6062 | 0.48 | 250 | 1.5494 |
1.5171 | 0.53 | 275 | 1.5284 |
1.6484 | 0.58 | 300 | 1.5091 |
1.7207 | 0.62 | 325 | 1.4958 |
1.6548 | 0.67 | 350 | 1.4817 |
1.6447 | 0.72 | 375 | 1.4746 |
1.5294 | 0.77 | 400 | 1.4358 |
1.6865 | 0.82 | 425 | 1.4269 |
1.4704 | 0.87 | 450 | 1.3963 |
1.4935 | 0.91 | 475 | 1.3714 |
1.4714 | 0.96 | 500 | 1.3496 |
1.4913 | 1.01 | 525 | 1.3327 |
1.3627 | 1.06 | 550 | 1.3060 |
1.2748 | 1.11 | 575 | 1.2857 |
1.1856 | 1.15 | 600 | 1.2624 |
1.1102 | 1.2 | 625 | 1.2413 |
1.2375 | 1.25 | 650 | 1.2214 |
1.2421 | 1.3 | 675 | 1.1989 |
1.1946 | 1.35 | 700 | 1.1823 |
1.2389 | 1.39 | 725 | 1.1674 |
1.2961 | 1.44 | 750 | 1.1567 |
1.1831 | 1.49 | 775 | 1.1566 |
1.2144 | 1.54 | 800 | 1.1326 |
1.2881 | 1.59 | 825 | 1.1279 |
1.2584 | 1.63 | 850 | 1.1073 |
1.2837 | 1.68 | 875 | 1.0878 |
1.1251 | 1.73 | 900 | 1.0812 |
1.0938 | 1.78 | 925 | 1.0706 |
1.0304 | 1.83 | 950 | 1.0636 |
1.313 | 1.88 | 975 | 1.0676 |
1.2245 | 1.92 | 1000 | 1.0604 |
1.1293 | 1.97 | 1025 | 1.0554 |
Framework versions
- PEFT 0.8.2
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1
- Downloads last month
- 0
Model tree for Maaz911/mistral-Mistral-Finetune-1
Base model
mistralai/Mistral-7B-v0.1