kennethge123
/

sst5-bert-base-uncased-kd

Model card Files Files and versions Community

sst5-bert-base-uncased-kd / README.md

kennethge123's picture

Upload README.md with huggingface_hub

cd81518 verified 8 months ago

|

311 Bytes

metadata

language: en
license: mit
library_name: pytorch

Plainly Optimized Network

Dataset: BIGBENCH

Trainer Hyperparameters:

lr = 5e-05
per_device_batch_size = 2
gradient_accumulation_steps = 4
weight_decay = 1e-09
seed = 42

eval_loss	eval_accuracy	epoch
61.745	0.059	1.0