English
Joelito kapllan commited on
Commit
f571250
1 Parent(s): 9a969c0

The learning rate was not displayed as it should. (#1)

Browse files

- The learning rate was not displayed as it should. (dae0aa3daa7b10c54bb5e0ed596dd79b6419bb40)


Co-authored-by: kapllan <kapllan@users.noreply.huggingface.co>

Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -80,7 +80,7 @@ For further details see [Niklaus et al. 2023](https://arxiv.org/abs/2306.02069?u
80
  - batche size: 512 samples
81
  - Number of steps: 1M/500K for the base/large model
82
  - Warm-up steps for the first 5\% of the total training steps
83
- - Learning rate: (linearly increasing up to) $1e\!-\!4$
84
  - Word masking: increased 20/30\% masking rate for base/large models respectively
85
 
86
  ## Evaluation
 
80
  - batche size: 512 samples
81
  - Number of steps: 1M/500K for the base/large model
82
  - Warm-up steps for the first 5\% of the total training steps
83
+ - Learning rate: (linearly increasing up to) 1e-4
84
  - Word masking: increased 20/30\% masking rate for base/large models respectively
85
 
86
  ## Evaluation