|
--- |
|
license: apache-2.0 |
|
base_model: EleutherAI/pythia-2.8b |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: patricia-walters-darkmagenta |
|
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. --> |
|
|
|
# patricia-walters-darkmagenta |
|
|
|
This model is a fine-tuned version of [EleutherAI/pythia-2.8b](https://huggingface.co./EleutherAI/pythia-2.8b) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5059 |
|
- Accuracy: 0.7 |
|
|
|
## 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: 1e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 16 |
|
- seed: 24 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 32 |
|
- total_eval_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: constant_with_warmup |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- training_steps: 2500 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:------:|:----:|:---------------:|:--------:| |
|
| 0.4856 | 1.2490 | 400 | 0.3020 | 1.0 | |
|
| 0.0737 | 2.4980 | 800 | 0.4773 | 0.7 | |
|
| 0.0886 | 3.7471 | 1200 | 1.2119 | 0.9 | |
|
| 0.1489 | 4.9961 | 1600 | 0.5459 | 0.8 | |
|
| 0.0285 | 6.2451 | 2000 | 2.4004 | 0.7 | |
|
| 0.0338 | 7.4941 | 2400 | 0.5059 | 0.7 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.40.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
