lysandre HF staff commited on
Commit
c2913cc
1 Parent(s): f5e146e

Update dimensions

Browse files
Files changed (1) hide show
  1. README.md +8 -0
README.md CHANGED
@@ -39,6 +39,14 @@ This way, the model learns an inner representation of the English language that
39
  useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard
40
  classifier using the features produced by the BERT model as inputs.
41
 
 
 
 
 
 
 
 
 
42
  ## Intended uses & limitations
43
 
44
  You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to
 
39
  useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard
40
  classifier using the features produced by the BERT model as inputs.
41
 
42
+
43
+ This model has the following configuration:
44
+
45
+ - 24-layer
46
+ - 1024 hidden dimension
47
+ - 16 attention heads
48
+ - 336M parameters.
49
+
50
  ## Intended uses & limitations
51
 
52
  You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to