|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: EleutherAI/pythia-70m |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: pythia_70m_sft |
|
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. --> |
|
|
|
# pythia_70m_sft |
|
|
|
This model is a fine-tuned version of [EleutherAI/pythia-70m](https://huggingface.co/EleutherAI/pythia-70m) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.4023 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 2.8612 | 0.0889 | 100 | 2.7705 | |
|
| 2.7799 | 0.1778 | 200 | 2.7116 | |
|
| 2.754 | 0.2667 | 300 | 2.6827 | |
|
| 2.7035 | 0.3556 | 400 | 2.6482 | |
|
| 2.6667 | 0.4444 | 500 | 2.6296 | |
|
| 2.6568 | 0.5333 | 600 | 2.6048 | |
|
| 2.6233 | 0.6222 | 700 | 2.5862 | |
|
| 2.5956 | 0.7111 | 800 | 2.5790 | |
|
| 2.5635 | 0.8 | 900 | 2.5436 | |
|
| 2.5469 | 0.8889 | 1000 | 2.5445 | |
|
| 2.499 | 0.9778 | 1100 | 2.5155 | |
|
| 2.3677 | 1.0667 | 1200 | 2.5158 | |
|
| 2.326 | 1.1556 | 1300 | 2.5018 | |
|
| 2.3132 | 1.2444 | 1400 | 2.5023 | |
|
| 2.3401 | 1.3333 | 1500 | 2.5026 | |
|
| 2.3027 | 1.4222 | 1600 | 2.4953 | |
|
| 2.3223 | 1.5111 | 1700 | 2.4804 | |
|
| 2.3186 | 1.6 | 1800 | 2.4776 | |
|
| 2.3187 | 1.6889 | 1900 | 2.4709 | |
|
| 2.3102 | 1.7778 | 2000 | 2.4618 | |
|
| 2.3129 | 1.8667 | 2100 | 2.4526 | |
|
| 2.287 | 1.9556 | 2200 | 2.4456 | |
|
| 2.1946 | 2.0444 | 2300 | 2.4423 | |
|
| 2.1757 | 2.1333 | 2400 | 2.4408 | |
|
| 2.1308 | 2.2222 | 2500 | 2.4383 | |
|
| 2.1475 | 2.3111 | 2600 | 2.4283 | |
|
| 2.155 | 2.4 | 2700 | 2.4231 | |
|
| 2.1197 | 2.4889 | 2800 | 2.4222 | |
|
| 2.1225 | 2.5778 | 2900 | 2.4173 | |
|
| 2.1196 | 2.6667 | 3000 | 2.4118 | |
|
| 2.146 | 2.7556 | 3100 | 2.4075 | |
|
| 2.1361 | 2.8444 | 3200 | 2.4046 | |
|
| 2.0965 | 2.9333 | 3300 | 2.4023 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.2 |
|
- Pytorch 2.5.1+cu121 |
|
- Datasets 3.1.0 |
|
- Tokenizers 0.20.3 |
|
|
