data-archetype commited on
Commit
7f3cd7c
·
verified ·
1 Parent(s): 77733fb

Add parameter count to technical report

Browse files
technical_report_capacitor_decoder.md CHANGED
@@ -45,6 +45,7 @@ configuration uses:
45
  | LR schedule | constant after warmup |
46
  | Warmup steps | `2,000` |
47
  | Batch size | `128` |
 
48
  | Gradient accumulation | `1` |
49
  | Gradient clip | `1.0` max norm |
50
  | Precision | AMP bfloat16 |
 
45
  | LR schedule | constant after warmup |
46
  | Warmup steps | `2,000` |
47
  | Batch size | `128` |
48
+ | Parameters | `61.93M` |
49
  | Gradient accumulation | `1` |
50
  | Gradient clip | `1.0` max norm |
51
  | Precision | AMP bfloat16 |