metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-v0.1
tags:
- generated_from_trainer
model-index:
- name: mistral-convsearch-7b-v4
results: []
mistral-convsearch-7b-v4
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: 0.4812
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: 4
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- total_eval_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5117 | 0.63 | 137 | 0.5122 |
| 0.4733 | 1.63 | 275 | 0.4869 |
| 0.4585 | 2.63 | 413 | 0.4761 |
| 0.4408 | 3.63 | 551 | 0.4705 |
| 0.4266 | 4.63 | 689 | 0.4692 |
| 0.4087 | 5.63 | 827 | 0.4686 |
| 0.399 | 6.63 | 965 | 0.4724 |
| 0.3875 | 7.63 | 1103 | 0.4756 |
| 0.3832 | 8.63 | 1240 | 0.4780 |
| 0.3682 | 9.63 | 1378 | 0.4814 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.1+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1