|
|
--- |
|
|
library_name: peft |
|
|
license: mit |
|
|
base_model: gpt2-medium |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: Se124M100KInfPrompt_WT_EOS_medium |
|
|
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. --> |
|
|
|
|
|
# Se124M100KInfPrompt_WT_EOS_medium |
|
|
|
|
|
This model is a fine-tuned version of [gpt2-medium](https://huggingface.co/gpt2-medium) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.7127 |
|
|
|
|
|
## 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: 64 |
|
|
- eval_batch_size: 64 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 4 |
|
|
- total_train_batch_size: 256 |
|
|
- 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_ratio: 0.03 |
|
|
- num_epochs: 3 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 2.8652 | 0.0655 | 20 | 2.6742 | |
|
|
| 2.6735 | 0.1309 | 40 | 2.4205 | |
|
|
| 2.3498 | 0.1964 | 60 | 2.0554 | |
|
|
| 1.9542 | 0.2619 | 80 | 1.6239 | |
|
|
| 1.5661 | 0.3273 | 100 | 1.2791 | |
|
|
| 1.3052 | 0.3928 | 120 | 1.0776 | |
|
|
| 1.1291 | 0.4583 | 140 | 0.9537 | |
|
|
| 1.0151 | 0.5237 | 160 | 0.8837 | |
|
|
| 0.9431 | 0.5892 | 180 | 0.8324 | |
|
|
| 0.8821 | 0.6547 | 200 | 0.8044 | |
|
|
| 0.8536 | 0.7201 | 220 | 0.7846 | |
|
|
| 0.8371 | 0.7856 | 240 | 0.7712 | |
|
|
| 0.8281 | 0.8511 | 260 | 0.7628 | |
|
|
| 0.8077 | 0.9165 | 280 | 0.7553 | |
|
|
| 0.8013 | 0.9820 | 300 | 0.7501 | |
|
|
| 0.7948 | 1.0458 | 320 | 0.7447 | |
|
|
| 0.783 | 1.1113 | 340 | 0.7394 | |
|
|
| 0.7727 | 1.1768 | 360 | 0.7372 | |
|
|
| 0.777 | 1.2422 | 380 | 0.7331 | |
|
|
| 0.7711 | 1.3077 | 400 | 0.7309 | |
|
|
| 0.7642 | 1.3732 | 420 | 0.7289 | |
|
|
| 0.7631 | 1.4386 | 440 | 0.7267 | |
|
|
| 0.7581 | 1.5041 | 460 | 0.7250 | |
|
|
| 0.7606 | 1.5696 | 480 | 0.7233 | |
|
|
| 0.7578 | 1.6350 | 500 | 0.7223 | |
|
|
| 0.7562 | 1.7005 | 520 | 0.7208 | |
|
|
| 0.7497 | 1.7660 | 540 | 0.7195 | |
|
|
| 0.7508 | 1.8314 | 560 | 0.7179 | |
|
|
| 0.7476 | 1.8969 | 580 | 0.7168 | |
|
|
| 0.7503 | 1.9624 | 600 | 0.7165 | |
|
|
| 0.7414 | 2.0262 | 620 | 0.7164 | |
|
|
| 0.7425 | 2.0917 | 640 | 0.7159 | |
|
|
| 0.7451 | 2.1571 | 660 | 0.7146 | |
|
|
| 0.7452 | 2.2226 | 680 | 0.7147 | |
|
|
| 0.7446 | 2.2881 | 700 | 0.7138 | |
|
|
| 0.7437 | 2.3535 | 720 | 0.7140 | |
|
|
| 0.7397 | 2.4190 | 740 | 0.7131 | |
|
|
| 0.7426 | 2.4845 | 760 | 0.7130 | |
|
|
| 0.7421 | 2.5499 | 780 | 0.7127 | |
|
|
| 0.7408 | 2.6154 | 800 | 0.7135 | |
|
|
| 0.7413 | 2.6809 | 820 | 0.7135 | |
|
|
| 0.7404 | 2.7463 | 840 | 0.7131 | |
|
|
| 0.7373 | 2.8118 | 860 | 0.7128 | |
|
|
| 0.7451 | 2.8773 | 880 | 0.7134 | |
|
|
| 0.7407 | 2.9427 | 900 | 0.7127 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.15.1 |
|
|
- Transformers 4.51.3 |
|
|
- Pytorch 2.6.0+cu118 |
|
|
- Datasets 3.5.0 |
|
|
- Tokenizers 0.21.1 |
