metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-Instruct-v0.1
tags:
- trl
- sft
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: original_glue_cola
results: []
original_glue_cola
This model is a fine-tuned version of mistralai/Mistral-7B-Instruct-v0.1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3803
- Accuracy: 0.8363
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: 16
- seed: 0
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.4253 | 0.22 | 50 | 0.4968 | 0.7553 |
| 0.5036 | 0.44 | 100 | 0.4704 | 0.7779 |
| 0.4974 | 0.66 | 150 | 0.4562 | 0.7825 |
| 0.4649 | 0.88 | 200 | 0.4299 | 0.7880 |
| 0.3356 | 1.1 | 250 | 0.4155 | 0.8051 |
| 0.4005 | 1.32 | 300 | 0.4026 | 0.8184 |
| 0.328 | 1.54 | 350 | 0.4052 | 0.8145 |
| 0.3632 | 1.76 | 400 | 0.3889 | 0.8270 |
| 0.3334 | 1.98 | 450 | 0.4176 | 0.8036 |
| 0.3166 | 2.2 | 500 | 0.4195 | 0.8324 |
| 0.2649 | 2.42 | 550 | 0.3929 | 0.8254 |
| 0.2805 | 2.64 | 600 | 0.3877 | 0.8363 |
| 0.3357 | 2.86 | 650 | 0.3734 | 0.8457 |
| 0.2476 | 3.08 | 700 | 0.3930 | 0.8418 |
| 0.2361 | 3.3 | 750 | 0.3893 | 0.8566 |
| 0.2375 | 3.52 | 800 | 0.3769 | 0.8519 |
| 0.18 | 3.74 | 850 | 0.3757 | 0.8519 |
| 0.2305 | 3.96 | 900 | 0.3783 | 0.8465 |
| 0.1432 | 4.18 | 950 | 0.4876 | 0.8402 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.1+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0