demerzel-iv's picture
upload model
a96a9b3
|
raw
history blame
573 Bytes
metadata
license: mit
language:
  - en
  - zh

Model Card for sparsing-law-0.1b-relu

This model is ReLU-activated and contains approximately 0.1 billion non-embedding parameters.

The model was trained from scratch using the pre-training dataset described in our paper, with the WSD (Warmup-Stable-Decay) learning rate scheduler. It represents the final checkpoint of the stable stage in WSD, meaning it has not undergone the decay stage.