mini_130m_2 / README.md
KaiyueWen's picture
Upload folder using huggingface_hub
a2139fc verified

Model Card

Best configuration

Hyperparameter Value
beta1 0.9
beta2 0.98
epsilon 1e-20
learning_rate 0.008
max_grad_norm 2
min_lr_ratio 0
nesterov False
train_batch_size 128
warmup 2000
weight_decay 0.1