|
--- |
|
license: mit |
|
base_model: gpt2 |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: codeparrot-ds |
|
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. --> |
|
|
|
# codeparrot-ds |
|
|
|
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.0621 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 256 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 1000 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 3.2102 | 0.02 | 1000 | 2.7478 | |
|
| 2.359 | 0.03 | 2000 | 2.2031 | |
|
| 2.0974 | 0.05 | 3000 | 1.9751 | |
|
| 1.9383 | 0.06 | 4000 | 1.8321 | |
|
| 1.8346 | 0.08 | 5000 | 1.7406 | |
|
| 1.7547 | 0.09 | 6000 | 1.6731 | |
|
| 1.6994 | 0.11 | 7000 | 1.6212 | |
|
| 1.6632 | 0.12 | 8000 | 1.5842 | |
|
| 1.6237 | 0.14 | 9000 | 1.5506 | |
|
| 1.5986 | 0.15 | 10000 | 1.5247 | |
|
| 1.5749 | 0.17 | 11000 | 1.4994 | |
|
| 1.5466 | 0.18 | 12000 | 1.4783 | |
|
| 1.5254 | 0.2 | 13000 | 1.4579 | |
|
| 1.5085 | 0.21 | 14000 | 1.4420 | |
|
| 1.4884 | 0.23 | 15000 | 1.4235 | |
|
| 1.4842 | 0.25 | 16000 | 1.4088 | |
|
| 1.4618 | 0.26 | 17000 | 1.3957 | |
|
| 1.4479 | 0.28 | 18000 | 1.3825 | |
|
| 1.4376 | 0.29 | 19000 | 1.3716 | |
|
| 1.4225 | 0.31 | 20000 | 1.3583 | |
|
| 1.4151 | 0.32 | 21000 | 1.3476 | |
|
| 1.4021 | 0.34 | 22000 | 1.3359 | |
|
| 1.3956 | 0.35 | 23000 | 1.3245 | |
|
| 1.3839 | 0.37 | 24000 | 1.3159 | |
|
| 1.3741 | 0.38 | 25000 | 1.3060 | |
|
| 1.3635 | 0.4 | 26000 | 1.2950 | |
|
| 1.3491 | 0.41 | 27000 | 1.2844 | |
|
| 1.3462 | 0.43 | 28000 | 1.2760 | |
|
| 1.3317 | 0.44 | 29000 | 1.2676 | |
|
| 1.3249 | 0.46 | 30000 | 1.2584 | |
|
| 1.3164 | 0.48 | 31000 | 1.2486 | |
|
| 1.3055 | 0.49 | 32000 | 1.2406 | |
|
| 1.3006 | 0.51 | 33000 | 1.2327 | |
|
| 1.2906 | 0.52 | 34000 | 1.2225 | |
|
| 1.2821 | 0.54 | 35000 | 1.2135 | |
|
| 1.2677 | 0.55 | 36000 | 1.2068 | |
|
| 1.2562 | 0.57 | 37000 | 1.1981 | |
|
| 1.2541 | 0.58 | 38000 | 1.1896 | |
|
| 1.2377 | 0.6 | 39000 | 1.1814 | |
|
| 1.2346 | 0.61 | 40000 | 1.1726 | |
|
| 1.2251 | 0.63 | 41000 | 1.1647 | |
|
| 1.2175 | 0.64 | 42000 | 1.1575 | |
|
| 1.2112 | 0.66 | 43000 | 1.1486 | |
|
| 1.2021 | 0.67 | 44000 | 1.1410 | |
|
| 1.1888 | 0.69 | 45000 | 1.1339 | |
|
| 1.1939 | 0.71 | 46000 | 1.1259 | |
|
| 1.18 | 0.72 | 47000 | 1.1198 | |
|
| 1.1698 | 0.74 | 48000 | 1.1130 | |
|
| 1.1634 | 0.75 | 49000 | 1.1063 | |
|
| 1.1593 | 0.77 | 50000 | 1.1006 | |
|
| 1.1545 | 0.78 | 51000 | 1.0946 | |
|
| 1.1478 | 0.8 | 52000 | 1.0896 | |
|
| 1.1443 | 0.81 | 53000 | 1.0855 | |
|
| 1.1365 | 0.83 | 54000 | 1.0808 | |
|
| 1.1332 | 0.84 | 55000 | 1.0773 | |
|
| 1.1336 | 0.86 | 56000 | 1.0736 | |
|
| 1.1276 | 0.87 | 57000 | 1.0711 | |
|
| 1.1241 | 0.89 | 58000 | 1.0686 | |
|
| 1.123 | 0.9 | 59000 | 1.0665 | |
|
| 1.1187 | 0.92 | 60000 | 1.0647 | |
|
| 1.1123 | 0.93 | 61000 | 1.0636 | |
|
| 1.1159 | 0.95 | 62000 | 1.0628 | |
|
| 1.1133 | 0.97 | 63000 | 1.0623 | |
|
| 1.1181 | 0.98 | 64000 | 1.0621 | |
|
| 1.1125 | 1.0 | 65000 | 1.0621 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.33.0.dev0 |
|
- Pytorch 2.0.0+cu117 |
|
- Datasets 2.14.4 |
|
- Tokenizers 0.13.3 |
|
|