Update README.md
Browse files
README.md
CHANGED
|
@@ -115,9 +115,11 @@ This model is not capable of generating grammatically or syntactically correct B
|
|
| 115 |
The model was evaluated on the validation split of the ai4bharat/naamapadam dataset to measure how well it models Bengali text.
|
| 116 |
|
| 117 |
## Metric: Perplexity (Lower is Better)
|
| 118 |
-
Model
|
| 119 |
-
|
| 120 |
-
|
|
|
|
|
|
|
| 121 |
|
| 122 |
## AddaGPT 2.0 shows a significantly lower perplexity, indicating a better fit to Bengali text.
|
| 123 |
|
|
|
|
| 115 |
The model was evaluated on the validation split of the ai4bharat/naamapadam dataset to measure how well it models Bengali text.
|
| 116 |
|
| 117 |
## Metric: Perplexity (Lower is Better)
|
| 118 |
+
| Model | Validation Perplexity |
|
| 119 |
+
| ----------------------- | --------------------- |
|
| 120 |
+
| **AddaGPT 2.0** | **25.61** |
|
| 121 |
+
| Vanilla GPT-2 (English) | 144.53 |
|
| 122 |
+
|
| 123 |
|
| 124 |
## AddaGPT 2.0 shows a significantly lower perplexity, indicating a better fit to Bengali text.
|
| 125 |
|