duccd commited on
Commit
e3d1eb6
·
verified ·
1 Parent(s): e2649a7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -3
README.md CHANGED
@@ -36,13 +36,15 @@ This work investigates two subtasks in temporal reasoning: 1. Date Arithmetic (d
36
 
37
  40,000 synthetic samples
38
 
39
- [More Information Needed]
40
-
41
  ### Training Procedure
42
 
43
  #### Training Hyperparameters
44
 
45
- - **Training regime:** fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision
 
 
 
 
46
 
47
  ## Evaluation
48
 
 
36
 
37
  40,000 synthetic samples
38
 
 
 
39
  ### Training Procedure
40
 
41
  #### Training Hyperparameters
42
 
43
+ - **Precision:** BF16
44
+ - **Learning rate:** 5.0e-5
45
+ - **Batch size per device:** 16
46
+ - **Epoch:** 5
47
+ - **Cutoff length:** 2048
48
 
49
  ## Evaluation
50