cope-g2b-2c-hs-skr-s1.5.9-sx-sk-s1.5.l1e4-e10-d25
This model is a fine-tuned version of google/gemma-2b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5312
- F1: 0.8875
- Precision: 0.8931
- Recall: 0.8820
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 5
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | F1 | Precision | Recall |
---|---|---|---|---|---|---|
0.0577 | 1.0 | 1075 | 0.2165 | 0.8375 | 0.7525 | 0.9441 |
0.106 | 2.0 | 2151 | 0.1881 | 0.8716 | 0.8391 | 0.9068 |
0.0097 | 3.0 | 3226 | 0.3158 | 0.8660 | 0.8688 | 0.8634 |
0.0002 | 4.0 | 4302 | 0.4903 | 0.8534 | 0.8973 | 0.8137 |
0.0125 | 5.0 | 5377 | 0.5278 | 0.8615 | 0.8537 | 0.8696 |
0.0137 | 6.0 | 6453 | 0.4507 | 0.8761 | 0.8529 | 0.9006 |
0.0 | 7.0 | 7528 | 0.5803 | 0.8807 | 0.8675 | 0.8944 |
0.0004 | 8.0 | 8604 | 0.5312 | 0.8875 | 0.8931 | 0.8820 |
0.0 | 9.0 | 9679 | 0.6352 | 0.8738 | 0.8659 | 0.8820 |
0.0 | 10.0 | 10750 | 0.6804 | 0.8738 | 0.8659 | 0.8820 |
Framework versions
- PEFT 0.8.2
- Transformers 4.38.0
- Pytorch 2.3.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2
- Downloads last month
- 38
Unable to determine this model’s pipeline type. Check the
docs
.