Update README.md
Browse files
README.md
CHANGED
|
@@ -17,6 +17,64 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
WitchLM is cool!
|
| 19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
### Training hyperparameters
|
| 21 |
|
| 22 |
<details>
|
|
|
|
| 17 |
|
| 18 |
WitchLM is cool!
|
| 19 |
|
| 20 |
+
## Benchmark
|
| 21 |
+
|
| 22 |
+
<details>
|
| 23 |
+
"leaderboard": {
|
| 24 |
+
"inst_level_strict_acc,none": 0.33573141486810554,
|
| 25 |
+
"inst_level_strict_acc_stderr,none": "N/A",
|
| 26 |
+
"inst_level_loose_acc,none": 0.39568345323741005,
|
| 27 |
+
"inst_level_loose_acc_stderr,none": "N/A",
|
| 28 |
+
"acc_norm,none": 0.3493319496692178,
|
| 29 |
+
"acc_norm_stderr,none": 0.005120138265236575,
|
| 30 |
+
"acc,none": 0.24418218085106383,
|
| 31 |
+
"acc_stderr,none": 0.003916649280281885,
|
| 32 |
+
"exact_match,none": 0.04078549848942598,
|
| 33 |
+
"exact_match_stderr,none": 0.005354025092648956,
|
| 34 |
+
"prompt_level_strict_acc,none": 0.1977818853974122,
|
| 35 |
+
"prompt_level_strict_acc_stderr,none": 0.01714125471908492,
|
| 36 |
+
"prompt_level_loose_acc,none": 0.25693160813308685,
|
| 37 |
+
"prompt_level_loose_acc_stderr,none": 0.018802962575636854,
|
| 38 |
+
"alias": "leaderboard"
|
| 39 |
+
},
|
| 40 |
+
"leaderboard_bbh": {
|
| 41 |
+
"acc_norm,none": 0.3591390383613956,
|
| 42 |
+
"acc_norm_stderr,none": 0.0058684522608536275,
|
| 43 |
+
"alias": " - leaderboard_bbh"
|
| 44 |
+
},
|
| 45 |
+
"leaderboard_gpqa": {
|
| 46 |
+
"acc_norm,none": 0.29194630872483224,
|
| 47 |
+
"acc_norm_stderr,none": 0.013178882651123217,
|
| 48 |
+
"alias": " - leaderboard_gpqa"
|
| 49 |
+
},
|
| 50 |
+
"leaderboard_ifeval": {
|
| 51 |
+
"prompt_level_strict_acc,none": 0.1977818853974122,
|
| 52 |
+
"prompt_level_strict_acc_stderr,none": 0.01714125471908492,
|
| 53 |
+
"inst_level_strict_acc,none": 0.33573141486810554,
|
| 54 |
+
"inst_level_strict_acc_stderr,none": "N/A",
|
| 55 |
+
"prompt_level_loose_acc,none": 0.25693160813308685,
|
| 56 |
+
"prompt_level_loose_acc_stderr,none": 0.018802962575636854,
|
| 57 |
+
"inst_level_loose_acc,none": 0.39568345323741005,
|
| 58 |
+
"inst_level_loose_acc_stderr,none": "N/A",
|
| 59 |
+
"alias": " - leaderboard_ifeval"
|
| 60 |
+
},
|
| 61 |
+
"leaderboard_math_hard": {
|
| 62 |
+
"exact_match,none": 0.04078549848942598,
|
| 63 |
+
"exact_match_stderr,none": 0.005354025092648956,
|
| 64 |
+
"alias": " - leaderboard_math_hard"
|
| 65 |
+
},
|
| 66 |
+
"leaderboard_mmlu_pro": {
|
| 67 |
+
"acc,none": 0.24418218085106383,
|
| 68 |
+
"acc_stderr,none": 0.003916649280281885,
|
| 69 |
+
"alias": " - leaderboard_mmlu_pro"
|
| 70 |
+
},
|
| 71 |
+
"leaderboard_musr": {
|
| 72 |
+
"acc_norm,none": 0.36507936507936506,
|
| 73 |
+
"acc_norm_stderr,none": 0.01715613678641816,
|
| 74 |
+
"alias": " - leaderboard_musr"
|
| 75 |
+
}
|
| 76 |
+
</details>
|
| 77 |
+
|
| 78 |
### Training hyperparameters
|
| 79 |
|
| 80 |
<details>
|