|
--- |
|
license: apache-2.0 |
|
base_model: Qwen/Qwen2-0.5B |
|
tags: |
|
- trl |
|
- sft |
|
- generated_from_trainer |
|
datasets: |
|
- generator |
|
model-index: |
|
- name: LawToken-0.5B-baseline |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# LawToken-0.5B-baseline |
|
|
|
This model is a fine-tuned version of [Qwen/Qwen2-0.5B](https://huggingface.co/Qwen/Qwen2-0.5B) on the generator dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8241 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 0.03 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 1.0668 | 0.13 | 10000 | 1.1084 | |
|
| 0.9744 | 0.27 | 20000 | 1.0197 | |
|
| 0.8864 | 0.4 | 30000 | 0.9631 | |
|
| 0.959 | 0.54 | 40000 | 0.9202 | |
|
| 0.8968 | 0.67 | 50000 | 0.8834 | |
|
| 0.8071 | 0.8 | 60000 | 0.8500 | |
|
| 0.7492 | 0.94 | 70000 | 0.8241 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.3.0a0+ebedce2 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.15.2 |
|
|
