metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-v0.1
tags:
- generated_from_trainer
model-index:
- name: Mistral_Sparse_refined_web_50p_2024-03-21
results: []
Mistral_Sparse_refined_web_50p_2024-03-21
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: 2.1512
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: 3
- gradient_accumulation_steps: 4
- total_train_batch_size: 12
- total_eval_batch_size: 3
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 501
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.4177 | 0.0 | 25 | 2.6401 |
| 2.5407 | 0.01 | 50 | 2.5820 |
| 2.3887 | 0.01 | 75 | 2.5299 |
| 2.2849 | 0.01 | 100 | 2.4991 |
| 2.2042 | 0.01 | 125 | 2.4802 |
| 2.2574 | 0.02 | 150 | 2.4609 |
| 2.2353 | 0.02 | 175 | 2.4473 |
| 2.3355 | 0.02 | 200 | 2.4449 |
| 2.3044 | 0.03 | 225 | 2.4381 |
| 2.2664 | 0.03 | 250 | 2.4348 |
| 2.1999 | 0.03 | 275 | 2.4263 |
| 2.2631 | 0.04 | 300 | 2.4247 |
| 2.2918 | 0.04 | 325 | 2.4184 |
| 2.1426 | 0.04 | 350 | 2.4185 |
| 2.149 | 0.04 | 375 | 2.4158 |
| 2.1937 | 0.05 | 400 | 2.4129 |
| 2.2372 | 0.05 | 425 | 2.4134 |
| 2.1997 | 0.05 | 450 | 2.4123 |
| 2.2937 | 0.06 | 475 | 2.4086 |
| 2.3067 | 0.06 | 500 | 2.4052 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0