Joelito kapllan commited on
Commit
3a4a2a7
1 Parent(s): 7d1665c

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

Browse files

- The learning rate was not displayed as it should. (ab91fb2cbe6ab1e43d946a73e9c9226d2d3fd000)


Co-authored-by: kapllan <[email protected]>

Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -82,7 +82,7 @@ For further details see [Niklaus et al. 2023](https://arxiv.org/abs/2306.02069?u
82
  - batche size: 512 samples
83
  - Number of steps: 1M/500K for the base/large model
84
  - Warm-up steps for the first 5\% of the total training steps
85
- - Learning rate: (linearly increasing up to) $1e\!-\!4$
86
  - Word masking: increased 20/30\% masking rate for base/large models respectively
87
 
88
  ## Evaluation
 
82
  - batche size: 512 samples
83
  - Number of steps: 1M/500K for the base/large model
84
  - Warm-up steps for the first 5\% of the total training steps
85
+ - Learning rate: (linearly increasing up to) 1e-4
86
  - Word masking: increased 20/30\% masking rate for base/large models respectively
87
 
88
  ## Evaluation