98 lines
3.5 KiB
Markdown
98 lines
3.5 KiB
Markdown
---
|
|
library_name: transformers
|
|
license: other
|
|
base_model: Qwen/Qwen2.5-0.5B
|
|
tags:
|
|
- llama-factory
|
|
- full
|
|
- generated_from_trainer
|
|
metrics:
|
|
- accuracy
|
|
model-index:
|
|
- name: qwen25_05b_base_full_ft_lunarlander_a4000
|
|
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. -->
|
|
|
|
# qwen25_05b_base_full_ft_lunarlander_a4000
|
|
|
|
This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B](https://huggingface.co/Qwen/Qwen2.5-0.5B) on the lunar_lander_270_reward_train dataset.
|
|
It achieves the following results on the evaluation set:
|
|
- Loss: 0.0253
|
|
- Accuracy: 0.9905
|
|
|
|
## 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: 1
|
|
- eval_batch_size: 1
|
|
- seed: 42
|
|
- distributed_type: multi-GPU
|
|
- gradient_accumulation_steps: 32
|
|
- total_train_batch_size: 32
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
|
- lr_scheduler_type: cosine
|
|
- lr_scheduler_warmup_steps: 100.0
|
|
- num_epochs: 1.0
|
|
- mixed_precision_training: Native AMP
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
|
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
|
| 0.1271 | 0.0327 | 200 | 0.1269 | 0.9408 |
|
|
| 0.1134 | 0.0653 | 400 | 0.1141 | 0.9496 |
|
|
| 0.0951 | 0.0980 | 600 | 0.0991 | 0.9562 |
|
|
| 0.0866 | 0.1306 | 800 | 0.0930 | 0.9564 |
|
|
| 0.0909 | 0.1633 | 1000 | 0.0883 | 0.9608 |
|
|
| 0.0980 | 0.1959 | 1200 | 0.0856 | 0.963 |
|
|
| 0.0836 | 0.2286 | 1400 | 0.0839 | 0.9611 |
|
|
| 0.0709 | 0.2612 | 1600 | 0.0811 | 0.9644 |
|
|
| 0.0757 | 0.2939 | 1800 | 0.0765 | 0.9665 |
|
|
| 0.0701 | 0.3265 | 2000 | 0.0791 | 0.9655 |
|
|
| 0.0717 | 0.3592 | 2200 | 0.0681 | 0.9706 |
|
|
| 0.0637 | 0.3918 | 2400 | 0.0704 | 0.9686 |
|
|
| 0.0619 | 0.4245 | 2600 | 0.0618 | 0.9747 |
|
|
| 0.0572 | 0.4571 | 2800 | 0.0562 | 0.9774 |
|
|
| 0.0605 | 0.4898 | 3000 | 0.0561 | 0.9767 |
|
|
| 0.0648 | 0.5224 | 3200 | 0.0584 | 0.9745 |
|
|
| 0.0659 | 0.5551 | 3400 | 0.0540 | 0.9771 |
|
|
| 0.0564 | 0.5878 | 3600 | 0.0479 | 0.9799 |
|
|
| 0.0545 | 0.6204 | 3800 | 0.0432 | 0.9819 |
|
|
| 0.0468 | 0.6531 | 4000 | 0.0449 | 0.9817 |
|
|
| 0.0388 | 0.6857 | 4200 | 0.0395 | 0.9840 |
|
|
| 0.0328 | 0.7184 | 4400 | 0.0397 | 0.9831 |
|
|
| 0.0363 | 0.7510 | 4600 | 0.0357 | 0.9856 |
|
|
| 0.0314 | 0.7837 | 4800 | 0.0362 | 0.985 |
|
|
| 0.0280 | 0.8163 | 5000 | 0.0336 | 0.9861 |
|
|
| 0.0308 | 0.8490 | 5200 | 0.0292 | 0.9882 |
|
|
| 0.0252 | 0.8816 | 5400 | 0.0268 | 0.9894 |
|
|
| 0.0202 | 0.9143 | 5600 | 0.0258 | 0.9901 |
|
|
| 0.0252 | 0.9469 | 5800 | 0.0255 | 0.9902 |
|
|
| 0.0201 | 0.9796 | 6000 | 0.0253 | 0.9904 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 5.2.0
|
|
- Pytorch 2.11.0+cu130
|
|
- Datasets 4.0.0
|
|
- Tokenizers 0.22.2
|