metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-v0.1
tags:
- generated_from_trainer
model-index:
- name: Mistral_Sparse_refined_web_resilu_2024-03-13
results: []
Mistral_Sparse_refined_web_resilu_2024-03-13
This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.5216
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: 1
- eval_batch_size: 1
- seed: 0
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 600
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
8.8267 | 0.0 | 25 | 8.7790 |
8.3017 | 0.01 | 50 | 8.3269 |
7.8921 | 0.01 | 75 | 7.9440 |
7.6064 | 0.02 | 100 | 7.6915 |
7.4736 | 0.02 | 125 | 7.5597 |
7.0663 | 0.02 | 150 | 6.8821 |
4.695 | 0.03 | 175 | 4.5616 |
3.6054 | 0.03 | 200 | 3.7193 |
3.2379 | 0.04 | 225 | 3.3697 |
3.0502 | 0.04 | 250 | 3.1769 |
2.8841 | 0.04 | 275 | 3.0628 |
2.8355 | 0.05 | 300 | 2.9948 |
2.794 | 0.05 | 325 | 2.9512 |
2.599 | 0.06 | 350 | 2.9101 |
2.7332 | 0.06 | 375 | 2.8858 |
2.762 | 0.06 | 400 | 2.8619 |
2.7191 | 0.07 | 425 | 2.8410 |
2.6786 | 0.07 | 450 | 2.8276 |
2.5627 | 0.08 | 475 | 2.8122 |
2.6268 | 0.08 | 500 | 2.8027 |
2.6987 | 0.08 | 525 | 2.7941 |
2.603 | 0.09 | 550 | 2.7820 |
2.7082 | 0.09 | 575 | 2.7705 |
2.5288 | 0.1 | 600 | 2.7647 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0