macadeliccc
commited on
Commit
•
f324ae3
1
Parent(s):
b7cd939
Update README.md
Browse files
README.md
CHANGED
@@ -48,4 +48,83 @@ The model is capable of quality code, math, and logical reasoning. Try whatever
|
|
48 |
|
49 |
# Evaluations
|
50 |
|
51 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
48 |
|
49 |
# Evaluations
|
50 |
|
51 |
+
| Model |AGIEval|GPT4All|TruthfulQA|Bigbench|Average|
|
52 |
+
|-------------------------------------------------------------------------|------:|------:|---------:|-------:|------:|
|
53 |
+
|[piccolo-math-2x7b](https://huggingface.co/macadeliccc/piccolo-math-2x7b)| 43.89| 74.98| 63.96| 44.99| 56.96|
|
54 |
+
|
55 |
+
### AGIEval
|
56 |
+
| Task |Version| Metric |Value| |Stderr|
|
57 |
+
|------------------------------|------:|--------|----:|---|-----:|
|
58 |
+
|agieval_aqua_rat | 0|acc |24.41|± | 2.70|
|
59 |
+
| | |acc_norm|24.80|± | 2.72|
|
60 |
+
|agieval_logiqa_en | 0|acc |35.79|± | 1.88|
|
61 |
+
| | |acc_norm|36.71|± | 1.89|
|
62 |
+
|agieval_lsat_ar | 0|acc |23.48|± | 2.80|
|
63 |
+
| | |acc_norm|23.91|± | 2.82|
|
64 |
+
|agieval_lsat_lr | 0|acc |49.22|± | 2.22|
|
65 |
+
| | |acc_norm|50.00|± | 2.22|
|
66 |
+
|agieval_lsat_rc | 0|acc |63.94|± | 2.93|
|
67 |
+
| | |acc_norm|64.31|± | 2.93|
|
68 |
+
|agieval_sat_en | 0|acc |77.18|± | 2.93|
|
69 |
+
| | |acc_norm|76.70|± | 2.95|
|
70 |
+
|agieval_sat_en_without_passage| 0|acc |45.15|± | 3.48|
|
71 |
+
| | |acc_norm|44.66|± | 3.47|
|
72 |
+
|agieval_sat_math | 0|acc |33.64|± | 3.19|
|
73 |
+
| | |acc_norm|30.00|± | 3.10|
|
74 |
+
|
75 |
+
Average: 43.89%
|
76 |
+
|
77 |
+
### GPT4All
|
78 |
+
| Task |Version| Metric |Value| |Stderr|
|
79 |
+
|-------------|------:|--------|----:|---|-----:|
|
80 |
+
|arc_challenge| 0|acc |61.86|± | 1.42|
|
81 |
+
| | |acc_norm|62.88|± | 1.41|
|
82 |
+
|arc_easy | 0|acc |84.34|± | 0.75|
|
83 |
+
| | |acc_norm|80.47|± | 0.81|
|
84 |
+
|boolq | 1|acc |86.88|± | 0.59|
|
85 |
+
|hellaswag | 0|acc |68.56|± | 0.46|
|
86 |
+
| | |acc_norm|85.16|± | 0.35|
|
87 |
+
|openbookqa | 0|acc |37.00|± | 2.16|
|
88 |
+
| | |acc_norm|47.80|± | 2.24|
|
89 |
+
|piqa | 0|acc |82.21|± | 0.89|
|
90 |
+
| | |acc_norm|83.68|± | 0.86|
|
91 |
+
|winogrande | 0|acc |77.98|± | 1.16|
|
92 |
+
|
93 |
+
Average: 74.98%
|
94 |
+
|
95 |
+
### TruthfulQA
|
96 |
+
| Task |Version|Metric|Value| |Stderr|
|
97 |
+
|-------------|------:|------|----:|---|-----:|
|
98 |
+
|truthfulqa_mc| 1|mc1 |47.37|± | 1.75|
|
99 |
+
| | |mc2 |63.96|± | 1.57|
|
100 |
+
|
101 |
+
Average: 63.96%
|
102 |
+
|
103 |
+
### Bigbench
|
104 |
+
| Task |Version| Metric |Value| |Stderr|
|
105 |
+
|------------------------------------------------|------:|---------------------|----:|---|-----:|
|
106 |
+
|bigbench_causal_judgement | 0|multiple_choice_grade|55.26|± | 3.62|
|
107 |
+
|bigbench_date_understanding | 0|multiple_choice_grade|63.14|± | 2.51|
|
108 |
+
|bigbench_disambiguation_qa | 0|multiple_choice_grade|42.64|± | 3.08|
|
109 |
+
|bigbench_geometric_shapes | 0|multiple_choice_grade|22.84|± | 2.22|
|
110 |
+
| | |exact_str_match | 3.34|± | 0.95|
|
111 |
+
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|36.60|± | 2.16|
|
112 |
+
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|25.57|± | 1.65|
|
113 |
+
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|56.00|± | 2.87|
|
114 |
+
|bigbench_movie_recommendation | 0|multiple_choice_grade|42.40|± | 2.21|
|
115 |
+
|bigbench_navigate | 0|multiple_choice_grade|54.70|± | 1.57|
|
116 |
+
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|62.90|± | 1.08|
|
117 |
+
|bigbench_ruin_names | 0|multiple_choice_grade|53.35|± | 2.36|
|
118 |
+
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|24.35|± | 1.36|
|
119 |
+
|bigbench_snarks | 0|multiple_choice_grade|62.43|± | 3.61|
|
120 |
+
|bigbench_sports_understanding | 0|multiple_choice_grade|70.28|± | 1.46|
|
121 |
+
|bigbench_temporal_sequences | 0|multiple_choice_grade|41.30|± | 1.56|
|
122 |
+
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|22.32|± | 1.18|
|
123 |
+
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|17.77|± | 0.91|
|
124 |
+
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|56.00|± | 2.87|
|
125 |
+
|
126 |
+
Average: 44.99%
|
127 |
+
|
128 |
+
Average score: 56.96%
|
129 |
+
|
130 |
+
Elapsed time: 01:51:53
|