sft-Mistral-7B-v0.2-on-memory_dialoges
This model is a fine-tuned version of alpindale/Mistral-7B-v0.2-hf on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5401
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: 5e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 10
- total_train_batch_size: 10
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 50
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
1.9835 | 0.03 | 20 | 1.9448 |
1.838 | 0.06 | 40 | 1.7001 |
1.4006 | 0.09 | 60 | 1.0107 |
0.8016 | 0.12 | 80 | 0.6707 |
0.6601 | 0.15 | 100 | 0.6268 |
0.6149 | 0.18 | 120 | 0.6046 |
0.6191 | 0.21 | 140 | 0.5913 |
0.6033 | 0.24 | 160 | 0.5777 |
0.5993 | 0.27 | 180 | 0.5671 |
0.5459 | 0.3 | 200 | 0.5572 |
0.5606 | 0.33 | 220 | 0.5528 |
0.5707 | 0.36 | 240 | 0.5500 |
0.5594 | 0.39 | 260 | 0.5478 |
0.5701 | 0.42 | 280 | 0.5460 |
0.5499 | 0.45 | 300 | 0.5446 |
0.5455 | 0.48 | 320 | 0.5437 |
0.5652 | 0.51 | 340 | 0.5426 |
0.5543 | 0.55 | 360 | 0.5418 |
0.5552 | 0.58 | 380 | 0.5412 |
0.5645 | 0.61 | 400 | 0.5408 |
0.5358 | 0.64 | 420 | 0.5404 |
0.55 | 0.67 | 440 | 0.5402 |
0.5604 | 0.7 | 460 | 0.5401 |
0.5686 | 0.73 | 480 | 0.5401 |
0.5509 | 0.76 | 500 | 0.5401 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2
- Datasets 2.15.0
- Tokenizers 0.15.1
- Downloads last month
- 1
Unable to determine this model’s pipeline type. Check the
docs
.