Commit
•
abbf091
1
Parent(s):
e8a68e5
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|
2 |
+
|----------------------------------------------------------------------|------:|------:|---------:|-------:|------:|
|
3 |
+
|[Lelantos-DPO-7B](https://huggingface.co/SanjiWatsuki/Lelantos-DPO-7B)| 45.47| 75| 67.05| 46.64| 58.54|
|
4 |
+
|
5 |
+
### AGIEval
|
6 |
+
| Task |Version| Metric |Value| |Stderr|
|
7 |
+
|------------------------------|------:|--------|----:|---|-----:|
|
8 |
+
|agieval_aqua_rat | 0|acc |25.20|± | 2.73|
|
9 |
+
| | |acc_norm|24.02|± | 2.69|
|
10 |
+
|agieval_logiqa_en | 0|acc |40.71|± | 1.93|
|
11 |
+
| | |acc_norm|40.25|± | 1.92|
|
12 |
+
|agieval_lsat_ar | 0|acc |24.35|± | 2.84|
|
13 |
+
| | |acc_norm|23.04|± | 2.78|
|
14 |
+
|agieval_lsat_lr | 0|acc |55.69|± | 2.20|
|
15 |
+
| | |acc_norm|55.49|± | 2.20|
|
16 |
+
|agieval_lsat_rc | 0|acc |65.06|± | 2.91|
|
17 |
+
| | |acc_norm|65.43|± | 2.91|
|
18 |
+
|agieval_sat_en | 0|acc |76.70|± | 2.95|
|
19 |
+
| | |acc_norm|76.70|± | 2.95|
|
20 |
+
|agieval_sat_en_without_passage| 0|acc |47.09|± | 3.49|
|
21 |
+
| | |acc_norm|45.63|± | 3.48|
|
22 |
+
|agieval_sat_math | 0|acc |36.36|± | 3.25|
|
23 |
+
| | |acc_norm|33.18|± | 3.18|
|
24 |
+
|
25 |
+
Average: 45.47%
|
26 |
+
|
27 |
+
### GPT4All
|
28 |
+
| Task |Version| Metric |Value| |Stderr|
|
29 |
+
|-------------|------:|--------|----:|---|-----:|
|
30 |
+
|arc_challenge| 0|acc |62.12|± | 1.42|
|
31 |
+
| | |acc_norm|63.23|± | 1.41|
|
32 |
+
|arc_easy | 0|acc |85.40|± | 0.72|
|
33 |
+
| | |acc_norm|81.02|± | 0.80|
|
34 |
+
|boolq | 1|acc |87.25|± | 0.58|
|
35 |
+
|hellaswag | 0|acc |67.97|± | 0.47|
|
36 |
+
| | |acc_norm|85.48|± | 0.35|
|
37 |
+
|openbookqa | 0|acc |36.80|± | 2.16|
|
38 |
+
| | |acc_norm|47.20|± | 2.23|
|
39 |
+
|piqa | 0|acc |81.88|± | 0.90|
|
40 |
+
| | |acc_norm|83.57|± | 0.86|
|
41 |
+
|winogrande | 0|acc |77.27|± | 1.18|
|
42 |
+
|
43 |
+
Average: 75.0%
|
44 |
+
|
45 |
+
### TruthfulQA
|
46 |
+
| Task |Version|Metric|Value| |Stderr|
|
47 |
+
|-------------|------:|------|----:|---|-----:|
|
48 |
+
|truthfulqa_mc| 1|mc1 |49.94|± | 1.75|
|
49 |
+
| | |mc2 |67.05|± | 1.53|
|
50 |
+
|
51 |
+
Average: 67.05%
|
52 |
+
|
53 |
+
### Bigbench
|
54 |
+
| Task |Version| Metric |Value| |Stderr|
|
55 |
+
|------------------------------------------------|------:|---------------------|----:|---|-----:|
|
56 |
+
|bigbench_causal_judgement | 0|multiple_choice_grade|58.95|± | 3.58|
|
57 |
+
|bigbench_date_understanding | 0|multiple_choice_grade|64.23|± | 2.50|
|
58 |
+
|bigbench_disambiguation_qa | 0|multiple_choice_grade|36.43|± | 3.00|
|
59 |
+
|bigbench_geometric_shapes | 0|multiple_choice_grade|23.68|± | 2.25|
|
60 |
+
| | |exact_str_match | 3.90|± | 1.02|
|
61 |
+
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|33.40|± | 2.11|
|
62 |
+
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|24.43|± | 1.63|
|
63 |
+
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|54.33|± | 2.88|
|
64 |
+
|bigbench_movie_recommendation | 0|multiple_choice_grade|52.20|± | 2.24|
|
65 |
+
|bigbench_navigate | 0|multiple_choice_grade|52.70|± | 1.58|
|
66 |
+
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|69.65|± | 1.03|
|
67 |
+
|bigbench_ruin_names | 0|multiple_choice_grade|50.22|± | 2.36|
|
68 |
+
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|40.98|± | 1.56|
|
69 |
+
|bigbench_snarks | 0|multiple_choice_grade|72.38|± | 3.33|
|
70 |
+
|bigbench_sports_understanding | 0|multiple_choice_grade|73.23|± | 1.41|
|
71 |
+
|bigbench_temporal_sequences | 0|multiple_choice_grade|39.90|± | 1.55|
|
72 |
+
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|20.88|± | 1.15|
|
73 |
+
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|17.60|± | 0.91|
|
74 |
+
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|54.33|± | 2.88|
|
75 |
+
|
76 |
+
Average: 46.64%
|
77 |
+
|
78 |
+
Average score: 58.54%
|