Update README.md
Browse files
README.md
CHANGED
|
@@ -112,6 +112,15 @@ def group_texts(examples):
|
|
| 112 |
lm_datasets = tokenized.map(group_texts, batched=True)
|
| 113 |
```
|
| 114 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 115 |
---
|
| 116 |
## Training Run & Metrics
|
| 117 |
- **Total steps**: 21,081
|
|
|
|
| 112 |
lm_datasets = tokenized.map(group_texts, batched=True)
|
| 113 |
```
|
| 114 |
|
| 115 |
+
---
|
| 116 |
+
|
| 117 |
+
## Tokens
|
| 118 |
+
- **Number of sequences in train set**: 899,394
|
| 119 |
+
- **Tokens per step**: 65,536
|
| 120 |
+
- **Steps per epoch**: 7,026
|
| 121 |
+
- **Total steps**: 21,078
|
| 122 |
+
- **Total tokens processed**: 1,381,367,808
|
| 123 |
+
|
| 124 |
---
|
| 125 |
## Training Run & Metrics
|
| 126 |
- **Total steps**: 21,081
|