metadata
license: other
base_model: Qwen/Qwen1.5-1.8B
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: Qwen1.5_1.8B_scotus
results: []
Qwen1.5_1.8B_scotus
This model is a fine-tuned version of Qwen/Qwen1.5-1.8B on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.6403
- Accuracy: 0.515
- F1 Macro: 0.3569
- F1 Micro: 0.515
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-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 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: 3.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
---|---|---|---|---|---|---|
2.1746 | 0.32 | 50 | 2.2792 | 0.2693 | 0.1148 | 0.2693 |
1.5758 | 0.64 | 100 | 1.9309 | 0.41 | 0.2291 | 0.41 |
1.7312 | 0.96 | 150 | 1.7475 | 0.4714 | 0.2547 | 0.4714 |
1.2859 | 1.27 | 200 | 1.7939 | 0.4686 | 0.3019 | 0.4686 |
1.1971 | 1.59 | 250 | 1.7146 | 0.4957 | 0.3307 | 0.4957 |
1.1271 | 1.91 | 300 | 1.6403 | 0.515 | 0.3569 | 0.515 |
0.8436 | 2.23 | 350 | 1.7995 | 0.51 | 0.3717 | 0.51 |
0.7723 | 2.55 | 400 | 1.8401 | 0.5029 | 0.3723 | 0.5029 |
0.7211 | 2.87 | 450 | 1.8232 | 0.5179 | 0.3753 | 0.5179 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2