Update README.md
Browse files
README.md
CHANGED
|
@@ -9,6 +9,8 @@ tags:
|
|
| 9 |
license: mit
|
| 10 |
language:
|
| 11 |
- en
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
SA stands for Safely and aligned.
|
|
@@ -32,6 +34,7 @@ Our training dataset consists of approximately 24K unique problem-tests pairs co
|
|
| 32 |
- PrimeIntellect SYNTHETIC-1
|
| 33 |
- LiveCodeBench v5 (5/1/23-7/31/24)
|
| 34 |
|
|
|
|
| 35 |
## Training Recipe
|
| 36 |
|
| 37 |
Our training recipe relies on an improved version of GRPO (GRPO+) and iterative context lengthening, introduced in DeepScaleR.
|
|
@@ -103,6 +106,8 @@ This permissive license ensures that researchers, developers, and enthusiasts wo
|
|
| 103 |
- Our model is trained on top of [`DeepSeek-R1-Distill-Qwen-14B`](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B).
|
| 104 |
- Our work is done as part of [Berkeley Sky Computing Lab](https://skycomputing.berkeley.edu/) and [Berkeley AI Research](https://bair.berkeley.edu/).
|
| 105 |
|
|
|
|
|
|
|
| 106 |
## Citation
|
| 107 |
```bibtex
|
| 108 |
@misc{deepcoder2025,
|
|
@@ -113,6 +118,8 @@ This permissive license ensures that researchers, developers, and enthusiasts wo
|
|
| 113 |
year={2025}
|
| 114 |
}
|
| 115 |
|
|
|
|
|
|
|
| 116 |
# Uploaded model
|
| 117 |
|
| 118 |
- **Developed by:** EpistemeAI
|
|
|
|
| 9 |
license: mit
|
| 10 |
language:
|
| 11 |
- en
|
| 12 |
+
datasets:
|
| 13 |
+
- UCSC-VLAA/STAR-1
|
| 14 |
---
|
| 15 |
|
| 16 |
SA stands for Safely and aligned.
|
|
|
|
| 34 |
- PrimeIntellect SYNTHETIC-1
|
| 35 |
- LiveCodeBench v5 (5/1/23-7/31/24)
|
| 36 |
|
| 37 |
+
- STAR-1
|
| 38 |
## Training Recipe
|
| 39 |
|
| 40 |
Our training recipe relies on an improved version of GRPO (GRPO+) and iterative context lengthening, introduced in DeepScaleR.
|
|
|
|
| 106 |
- Our model is trained on top of [`DeepSeek-R1-Distill-Qwen-14B`](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B).
|
| 107 |
- Our work is done as part of [Berkeley Sky Computing Lab](https://skycomputing.berkeley.edu/) and [Berkeley AI Research](https://bair.berkeley.edu/).
|
| 108 |
|
| 109 |
+
- thanks to UCSC-VLAA
|
| 110 |
+
|
| 111 |
## Citation
|
| 112 |
```bibtex
|
| 113 |
@misc{deepcoder2025,
|
|
|
|
| 118 |
year={2025}
|
| 119 |
}
|
| 120 |
|
| 121 |
+
|
| 122 |
+
|
| 123 |
# Uploaded model
|
| 124 |
|
| 125 |
- **Developed by:** EpistemeAI
|