Update README.md
Browse files
README.md
CHANGED
|
@@ -50,21 +50,3 @@ seed: 42
|
|
| 50 |
warmup_ratio: 0.1
|
| 51 |
save_only_model: true
|
| 52 |
```
|
| 53 |
-
|
| 54 |
-
## Evaluation Results
|
| 55 |
-
| Datasets | Llama3.1-8B-Instruct | FuseChat-Llama-3.1-8B-SFT | FuseChat-Llama-3.1-8B-Instruct |
|
| 56 |
-
|---------------------------------|----------------------|---------------------------|--------------------------------|
|
| 57 |
-
| AlpacaEval-2 (LC/WR) | 28.3/28.7 | 41.3/37.7 | 65.4/63.3 |
|
| 58 |
-
| Arena-Hard (WR/SC) | 28.1/23.8 | 38.7/29 | 58.2/46.4 |
|
| 59 |
-
| MT-Bench | 8.38 | 8.54 | 9 |
|
| 60 |
-
| AlignBench v1.1 | 4.61 | 6.25 | 6.69 |
|
| 61 |
-
| LiveBench 0831 | 27.6 | 30.2 | 32 |
|
| 62 |
-
| GSM8K | 85.9 | 87 | 88 |
|
| 63 |
-
| MATH | 50.7 | 54.7 | 55.2 |
|
| 64 |
-
| AMC 23 | 25 | 30 | 37.5 |
|
| 65 |
-
| MMLU-Pro | 50 | 47.8 | 49.2 |
|
| 66 |
-
| MMLU-redux | 67.2 | 68.4 | 69.2 |
|
| 67 |
-
| GPQA-Diamond | 33.8 | 37.9 | 34.9 |
|
| 68 |
-
| HumanEval | 69.5 | 69.5 | 71.3 |
|
| 69 |
-
| MBPP | 75.4 | 71.4 | 72 |
|
| 70 |
-
| LiveCodeBench 2408-2411 (all/esay) | 12.3/40.5 | 12.6/39 | 13.1/43.2 |
|
|
|
|
| 50 |
warmup_ratio: 0.1
|
| 51 |
save_only_model: true
|
| 52 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|