datasciguy
commited on
Commit
•
1967817
1
Parent(s):
4801cca
Update README.md
Browse files
README.md
CHANGED
@@ -14,7 +14,7 @@ license: mit
|
|
14 |
- The dataset improves spelling, grammar, and consistency while replacing references to violent crimes with non-violent activities and removes self-censorship from explicatives.
|
15 |
|
16 |
**Training Time**: Approximately 30-45 minutes. Each validation epoch takes ~322 seconds.
|
17 |
-
**Hardware**: Trained on
|
18 |
|
19 |
---
|
20 |
|
@@ -54,7 +54,8 @@ license: mit
|
|
54 |
- Perplexity: 2.18
|
55 |
- Epoch completed in 322.01 seconds
|
56 |
|
57 |
-
-
|
|
|
58 |
- Training Loss: 0.2831
|
59 |
- Validation Loss: 0.8017
|
60 |
- Perplexity: 2.23
|
|
|
14 |
- The dataset improves spelling, grammar, and consistency while replacing references to violent crimes with non-violent activities and removes self-censorship from explicatives.
|
15 |
|
16 |
**Training Time**: Approximately 30-45 minutes. Each validation epoch takes ~322 seconds.
|
17 |
+
**Hardware**: Trained on Google Colab Pro A100 GPU (40GB).
|
18 |
|
19 |
---
|
20 |
|
|
|
54 |
- Perplexity: 2.18
|
55 |
- Epoch completed in 322.01 seconds
|
56 |
|
57 |
+
-
|
58 |
+
**Epoch 5**:
|
59 |
- Training Loss: 0.2831
|
60 |
- Validation Loss: 0.8017
|
61 |
- Perplexity: 2.23
|