gamepollakrit
commited on
Commit
•
830c3ce
1
Parent(s):
b468814
Update README.md
Browse files
README.md
CHANGED
@@ -28,6 +28,18 @@ pipeline_tag: text-generation
|
|
28 |
|
29 |
---
|
30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
### Prompt Template
|
32 |
|
33 |
This model uses `ChatML` prompt template:
|
|
|
28 |
|
29 |
---
|
30 |
|
31 |
+
### Performance Evaluation
|
32 |
+
|
33 |
+
Below are the benchmark results of Tsunami-1.0-14B-Instruct compared to similar models in its class:
|
34 |
+
|
35 |
+
| Model | Average | Thai Exam | M3Exam |
|
36 |
+
| --- | --- | --- | --- |
|
37 |
+
| Qwen2.5-14B-Instruct | 58.45 | 57.35 | 59.55 |
|
38 |
+
| Meta-Llama-3.1-70B-Instruct | 59.38 | 58.23 | 60.52 |
|
39 |
+
| llama-3-typhoon-v1.5x-70b-instruct | 59.34 | 58.76 | 59.92 |
|
40 |
+
| openthaigpt1.5-14b-instruct | 60.41 | 58.41 | 62.41 |
|
41 |
+
| **Tsunami-1.0-14B-Instruct** | **62.05** | **61.06** | **63.05** |
|
42 |
+
|
43 |
### Prompt Template
|
44 |
|
45 |
This model uses `ChatML` prompt template:
|