Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: transformers
|
| 4 |
+
---
|
| 5 |
+
# MyAwesomeModel
|
| 6 |
+
|
| 7 |
+
## Overview
|
| 8 |
+
|
| 9 |
+
MyAwesomeModel is a state-of-the-art language model trained with significant computational resources and algorithmic optimizations during post-training. The model demonstrates outstanding performance across various benchmark evaluations, including mathematics, programming, and general logic.
|
| 10 |
+
|
| 11 |
+
## Evaluation Results
|
| 12 |
+
|
| 13 |
+
The best checkpoint (step_1000) achieves the following scores across all 15 benchmarks:
|
| 14 |
+
|
| 15 |
+
| Category | Benchmark | Score |
|
| 16 |
+
|---|---|---|
|
| 17 |
+
| **Core Reasoning Tasks** | Math Reasoning | 0.55 |
|
| 18 |
+
| | Logical Reasoning | 0.819 |
|
| 19 |
+
| | Common Sense | 0.736 |
|
| 20 |
+
| **Language Understanding** | Reading Comprehension | 0.7 |
|
| 21 |
+
| | Question Answering | 0.607 |
|
| 22 |
+
| | Text Classification | 0.828 |
|
| 23 |
+
| | Sentiment Analysis | 0.792 |
|
| 24 |
+
| **Generation Tasks** | Code Generation | 0.65 |
|
| 25 |
+
| | Creative Writing | 0.61 |
|
| 26 |
+
| | Dialogue Generation | 0.644 |
|
| 27 |
+
| | Summarization | 0.767 |
|
| 28 |
+
| **Specialized Capabilities** | Translation | 0.804 |
|
| 29 |
+
| | Knowledge Retrieval | 0.676 |
|
| 30 |
+
| | Instruction Following | 0.758 |
|
| 31 |
+
| | Safety Evaluation | 0.739 |
|
| 32 |
+
|
| 33 |
+
## Overall Performance
|
| 34 |
+
|
| 35 |
+
- **Best checkpoint:** step_1000
|
| 36 |
+
- **Overall weighted score:** 0.71
|
| 37 |
+
|
| 38 |
+
Compared to the previous version, the upgraded model shows significant improvements in handling complex reasoning tasks. For instance, in the AIME 2025 test, the model's accuracy has increased from 70% in the previous version to 87.5% in the current version.
|
| 39 |
+
|
| 40 |
+
## Usage
|
| 41 |
+
|
| 42 |
+
We recommend setting the temperature parameter to 0.6.
|
| 43 |
+
|
| 44 |
+
### System Prompt
|
| 45 |
+
```
|
| 46 |
+
You are MyAwesomeModel, a helpful AI assistant.
|
| 47 |
+
Today is {current date}.
|
| 48 |
+
```
|
| 49 |
+
|
| 50 |
+
## License
|
| 51 |
+
This model is licensed under the MIT License. The model series supports commercial use and distillation.
|