Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: transformers
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# MyAwesomeModel-TestRepository
|
| 7 |
+
|
| 8 |
+
## Model Details
|
| 9 |
+
This is the best checkpoint (step_700) selected based on **highest eval_accuracy = 0.789** across all 10 checkpoints evaluated.
|
| 10 |
+
|
| 11 |
+
## Comprehensive Benchmark Evaluation Results
|
| 12 |
+
All scores reported to **three decimal places**:
|
| 13 |
+
|
| 14 |
+
| Benchmark Category | Score |
|
| 15 |
+
|---|---|
|
| 16 |
+
| **Core Reasoning Tasks** | |
|
| 17 |
+
| Math Reasoning | 0.792 |
|
| 18 |
+
| Logical Reasoning | 0.815 |
|
| 19 |
+
| Common Sense | 0.741 |
|
| 20 |
+
| **Language Understanding** | |
|
| 21 |
+
| Reading Comprehension | 0.703 |
|
| 22 |
+
| Question Answering | 0.628 |
|
| 23 |
+
| Text Classification | 0.837 |
|
| 24 |
+
| Sentiment Analysis | 0.806 |
|
| 25 |
+
| **Generation Tasks** | |
|
| 26 |
+
| Code Generation | 0.667 |
|
| 27 |
+
| Creative Writing | 0.619 |
|
| 28 |
+
| Dialogue Generation | 0.654 |
|
| 29 |
+
| Summarization | 0.778 |
|
| 30 |
+
| **Specialized Capabilities** | |
|
| 31 |
+
| Translation | 0.815 |
|
| 32 |
+
| Knowledge Retrieval | 0.685 |
|
| 33 |
+
| Instruction Following | 0.770 |
|
| 34 |
+
| Safety Evaluation | 0.743 |
|
| 35 |
+
|
| 36 |
+
### Overall Weighted Eval Accuracy: **0.789**
|