80 lines
2.3 KiB
Markdown
80 lines
2.3 KiB
Markdown
---
|
|
license: mit
|
|
tags:
|
|
- generated_from_trainer
|
|
datasets:
|
|
- generator
|
|
model-index:
|
|
- name: cbt-guten-rarity-all-mixed-cut-2p6k
|
|
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. -->
|
|
|
|
# cbt-guten-rarity-all-mixed-cut-2p6k
|
|
|
|
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
|
|
It achieves the following results on the evaluation set:
|
|
- Loss: 4.3210
|
|
|
|
## 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: 0.0005
|
|
- train_batch_size: 64
|
|
- eval_batch_size: 64
|
|
- seed: 42
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
- lr_scheduler_type: cosine
|
|
- lr_scheduler_warmup_steps: 1000
|
|
- num_epochs: 6
|
|
- mixed_precision_training: Native AMP
|
|
|
|
### Training results
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss |
|
|
|:-------------:|:-----:|:-----:|:---------------:|
|
|
| 6.6926 | 0.29 | 500 | 5.6317 |
|
|
| 5.3392 | 0.59 | 1000 | 5.1955 |
|
|
| 5.0004 | 0.88 | 1500 | 4.9509 |
|
|
| 4.7275 | 1.17 | 2000 | 4.8088 |
|
|
| 4.5644 | 1.46 | 2500 | 4.6844 |
|
|
| 4.4518 | 1.76 | 3000 | 4.5759 |
|
|
| 4.3336 | 2.05 | 3500 | 4.5021 |
|
|
| 4.137 | 2.34 | 4000 | 4.4540 |
|
|
| 4.1052 | 2.63 | 4500 | 4.3960 |
|
|
| 4.0754 | 2.93 | 5000 | 4.3380 |
|
|
| 3.8637 | 3.22 | 5500 | 4.3349 |
|
|
| 3.8081 | 3.51 | 6000 | 4.3053 |
|
|
| 3.7909 | 3.8 | 6500 | 4.2708 |
|
|
| 3.6948 | 4.1 | 7000 | 4.2703 |
|
|
| 3.5239 | 4.39 | 7500 | 4.2668 |
|
|
| 3.5181 | 4.68 | 8000 | 4.2517 |
|
|
| 3.5066 | 4.97 | 8500 | 4.2362 |
|
|
| 3.3432 | 5.27 | 9000 | 4.2494 |
|
|
| 3.3256 | 5.56 | 9500 | 4.2492 |
|
|
| 3.3315 | 5.85 | 10000 | 4.2485 |
|
|
|
|
|
|
### Framework versions
|
|
|
|
- Transformers 4.26.1
|
|
- Pytorch 1.11.0+cu113
|
|
- Datasets 2.13.0
|
|
- Tokenizers 0.13.3
|