|
|
--- |
|
|
library_name: peft |
|
|
license: apache-2.0 |
|
|
base_model: t5-base |
|
|
tags: |
|
|
- base_model:adapter:t5-base |
|
|
- lora |
|
|
- transformers |
|
|
model-index: |
|
|
- name: t5-base-controllable-qa-generator |
|
|
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. --> |
|
|
|
|
|
# t5-base-controllable-qa-generator |
|
|
|
|
|
This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.8687 |
|
|
|
|
|
## 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: 8 |
|
|
- eval_batch_size: 8 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 4 |
|
|
- total_train_batch_size: 32 |
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: linear |
|
|
- training_steps: 2000 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-------:|:----:|:---------------:| |
|
|
| 7.2453 | 2.3855 | 50 | 3.5197 | |
|
|
| 2.0995 | 4.7711 | 100 | 1.9805 | |
|
|
| 1.8109 | 7.1446 | 150 | 1.7060 | |
|
|
| 1.6062 | 9.5301 | 200 | 1.5023 | |
|
|
| 1.3582 | 11.9157 | 250 | 1.3483 | |
|
|
| 1.3196 | 14.2892 | 300 | 1.2469 | |
|
|
| 1.2131 | 16.6747 | 350 | 1.1814 | |
|
|
| 1.1552 | 19.0482 | 400 | 1.1317 | |
|
|
| 1.122 | 21.4337 | 450 | 1.0944 | |
|
|
| 1.046 | 23.8193 | 500 | 1.0690 | |
|
|
| 1.051 | 26.1928 | 550 | 1.0470 | |
|
|
| 1.0073 | 28.5783 | 600 | 1.0285 | |
|
|
| 0.9617 | 30.9639 | 650 | 1.0111 | |
|
|
| 0.9889 | 33.3373 | 700 | 0.9974 | |
|
|
| 0.9275 | 35.7229 | 750 | 0.9843 | |
|
|
| 0.923 | 38.0964 | 800 | 0.9723 | |
|
|
| 0.9385 | 40.4819 | 850 | 0.9607 | |
|
|
| 0.931 | 42.8675 | 900 | 0.9519 | |
|
|
| 0.8915 | 45.2410 | 950 | 0.9439 | |
|
|
| 0.939 | 47.6265 | 1000 | 0.9360 | |
|
|
| 0.9003 | 50.0 | 1050 | 0.9281 | |
|
|
| 0.875 | 52.3855 | 1100 | 0.9206 | |
|
|
| 0.8751 | 54.7711 | 1150 | 0.9152 | |
|
|
| 0.8616 | 57.1446 | 1200 | 0.9095 | |
|
|
| 0.8805 | 59.5301 | 1250 | 0.9040 | |
|
|
| 0.8509 | 61.9157 | 1300 | 0.8989 | |
|
|
| 0.8196 | 64.2892 | 1350 | 0.8942 | |
|
|
| 0.8495 | 66.6747 | 1400 | 0.8908 | |
|
|
| 0.8655 | 69.0482 | 1450 | 0.8874 | |
|
|
| 0.9224 | 71.4337 | 1500 | 0.8829 | |
|
|
| 0.8371 | 73.8193 | 1550 | 0.8807 | |
|
|
| 0.8107 | 76.1928 | 1600 | 0.8784 | |
|
|
| 0.8333 | 78.5783 | 1650 | 0.8756 | |
|
|
| 0.8475 | 80.9639 | 1700 | 0.8747 | |
|
|
| 0.8075 | 83.3373 | 1750 | 0.8729 | |
|
|
| 0.8352 | 85.7229 | 1800 | 0.8712 | |
|
|
| 0.8516 | 88.0964 | 1850 | 0.8700 | |
|
|
| 0.8115 | 90.4819 | 1900 | 0.8694 | |
|
|
| 0.8477 | 92.8675 | 1950 | 0.8687 | |
|
|
| 0.8289 | 95.2410 | 2000 | 0.8687 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.16.0 |
|
|
- Transformers 4.53.3 |
|
|
- Pytorch 2.6.0+cu124 |
|
|
- Datasets 4.1.1 |
|
|
- Tokenizers 0.21.2 |
