zirui3's picture
add config files for training
f5b6f8f
|
raw
history blame
1.1 kB
---
license: cc-by-4.0
---
**pythia-1.4B-finetuned-oa-instructions**
This model is a fine-tuned version of pythia on the oa dataset. It achieves the following results on the evaluation set:
Loss: 0.1224
**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:
* seed: 42
* learning_rate: 5e-06
* train_batch_size: 32
* eval_batch_size: 8
* optimizer: Adam with betas : {'lr': 5e-06, 'betas': [0.9, 0.999], 'eps': 1e-08, 'weight_decay': 0.0}
* lr_scheduler_type: linear
* training_steps: 5000
* fp16
* warmup_steps 5
* Num examples = 53k
**Training results**
```
{
"epoch": 1.0,
"train_loss": 0.8031303182039198,
"train_runtime": 6338.6403,
"train_samples": 53455,
"train_samples_per_second": 8.433,
"train_steps_per_second": 0.264
}
```
**Framework versions**
* transformers 4.24.0
* torch 1.10.0+cu111
* datasets 2.10.0
* tokenizers 0.12.1