---
language: en
license: mit
library_name: pytorch
---
# Plainly Optimized Network
Dataset: BIGBENCH

Trainer Hyperparameters:
- `lr` = 5e-05
- `per_device_batch_size` = 8
- `gradient_accumulation_steps` = 2
- `weight_decay` = 0.0
- `seed` = 42

|eval_loss|eval_accuracy|epoch|
|--|--|--|
|10.379|0.571|1.0|
|9.388|0.643|2.0|
|10.286|0.571|3.0|
|10.324|0.571|4.0|
|10.254|0.571|5.0|
|10.166|0.571|6.0|
|10.122|0.571|7.0|
|10.020|0.571|8.0|
|10.035|0.571|9.0|
|9.961|0.571|10.0|
|9.963|0.571|11.0|
|9.962|0.571|12.0|
|9.990|0.500|13.0|
|10.817|0.571|14.0|
|10.030|0.571|15.0|
|10.049|0.571|16.0|
|10.057|0.571|17.0|
|10.067|0.571|18.0|
|10.080|0.571|19.0|