ZeroShot-3.3.20-Mistral-7b-Multilanguage-3.2.0
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3347
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: 2
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.5234 | 0.06 | 100 | 0.4902 |
0.4606 | 0.12 | 200 | 0.4453 |
0.4223 | 0.19 | 300 | 0.4323 |
0.4124 | 0.25 | 400 | 0.4239 |
0.4023 | 0.31 | 500 | 0.4136 |
0.3954 | 0.37 | 600 | 0.4068 |
0.3835 | 0.43 | 700 | 0.4009 |
0.3815 | 0.5 | 800 | 0.3953 |
0.3738 | 0.56 | 900 | 0.3917 |
0.3728 | 0.62 | 1000 | 0.3870 |
0.3846 | 0.68 | 1100 | 0.3816 |
0.3872 | 0.74 | 1200 | 0.3773 |
0.3823 | 0.81 | 1300 | 0.3739 |
0.362 | 0.87 | 1400 | 0.3700 |
0.3661 | 0.93 | 1500 | 0.3663 |
0.3572 | 0.99 | 1600 | 0.3628 |
0.3185 | 1.05 | 1700 | 0.3611 |
0.3133 | 1.12 | 1800 | 0.3571 |
0.3153 | 1.18 | 1900 | 0.3538 |
0.3144 | 1.24 | 2000 | 0.3513 |
0.3082 | 1.3 | 2100 | 0.3480 |
0.315 | 1.36 | 2200 | 0.3460 |
0.3062 | 1.43 | 2300 | 0.3433 |
0.305 | 1.49 | 2400 | 0.3417 |
0.3131 | 1.55 | 2500 | 0.3396 |
0.3049 | 1.61 | 2600 | 0.3380 |
0.2971 | 1.67 | 2700 | 0.3371 |
0.2983 | 1.74 | 2800 | 0.3360 |
0.2987 | 1.8 | 2900 | 0.3353 |
0.2998 | 1.86 | 3000 | 0.3350 |
0.3066 | 1.92 | 3100 | 0.3348 |
0.3004 | 1.98 | 3200 | 0.3347 |
Framework versions
- PEFT 0.9.0
- Transformers 4.38.2
- Pytorch 2.1.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 6
Model tree for Weni/ZeroShot-3.3.20-Mistral-7b-Multilanguage-3.2.0
Base model
mistralai/Mistral-7B-Instruct-v0.2
Adapter
this model