metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: NickyNicky/Mistral-7B-OpenOrca-oasst_top1_2023-08-25-v2
model-index:
- name: mistral-sigoIAspirantes-Orca-oass-v2
results: []
mistral-sigoIAspirantes-Orca-oass-v2
This model is a fine-tuned version of NickyNicky/Mistral-7B-OpenOrca-oasst_top1_2023-08-25-v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1354
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: 2.5e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 1
- training_steps: 400
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.6909 | 0.35 | 25 | 1.2159 |
| 1.0275 | 0.7 | 50 | 0.8422 |
| 0.7147 | 1.06 | 75 | 0.6233 |
| 0.5131 | 1.41 | 100 | 0.5002 |
| 0.3872 | 1.76 | 125 | 0.3971 |
| 0.3419 | 2.11 | 150 | 0.3027 |
| 0.1742 | 2.46 | 175 | 0.2725 |
| 0.2616 | 2.82 | 200 | 0.2068 |
| 0.1427 | 3.17 | 225 | 0.2039 |
| 0.1228 | 3.52 | 250 | 0.1627 |
| 0.1186 | 3.87 | 275 | 0.1441 |
| 0.0751 | 4.23 | 300 | 0.1540 |
| 0.0892 | 4.58 | 325 | 0.1524 |
| 0.0833 | 4.93 | 350 | 0.1327 |
| 0.064 | 5.28 | 375 | 0.1347 |
| 0.06 | 5.63 | 400 | 0.1354 |
Framework versions
- PEFT 0.10.1.dev0
- Transformers 4.40.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2