Update README.md
Browse filesadded training tokens in total.
README.md
CHANGED
@@ -33,7 +33,7 @@ Compared to Argonne-1.0, this model features significant enhancements, including
|
|
33 |
|
34 |
## Training Data 📚
|
35 |
|
36 |
-
This model was trained on the **same dataset** used for Argonne-1.0 pretraining.
|
37 |
|
38 |
---
|
39 |
|
@@ -98,65 +98,65 @@ The meaning of life is tamed in many ways. It is a state of mental and physical
|
|
98 |
## **Overall Accuracy**
|
99 |
**0.2549 (3579/14042)**
|
100 |
|
101 |
-
## Breakdown by Subject:
|
102 |
| Subject | Accuracy (Correct/Total) |
|
103 |
|------------------------------|----------------------|
|
104 |
-
| international_law | 0.3636 (44/121) |
|
105 |
-
| anatomy | 0.3333 (45/135) |
|
106 |
| abstract_algebra | 0.3200 (32/100) |
|
107 |
-
|
|
108 |
-
|
|
|
|
|
|
109 |
| college_biology | 0.3056 (44/144) |
|
110 |
-
|
|
111 |
-
| high_school_chemistry | 0.3054 (62/203) |
|
112 |
-
| world_religions | 0.3041 (52/171) |
|
113 |
-
| prehistory | 0.3025 (98/324) |
|
114 |
-
| high_school_european_history | 0.2970 (49/165) |
|
115 |
-
| elementary_mathematics | 0.2963 (112/378) |
|
116 |
-
| logical_fallacies | 0.2945 (48/163) |
|
117 |
-
| computer_security | 0.2900 (29/100) |
|
118 |
-
| professional_psychology | 0.2827 (173/612) |
|
119 |
-
| moral_disputes | 0.2803 (97/346) |
|
120 |
-
| professional_accounting | 0.2766 (78/282) |
|
121 |
-
| electrical_engineering | 0.2759 (40/145) |
|
122 |
| college_computer_science | 0.2700 (27/100) |
|
|
|
123 |
| college_mathematics | 0.2700 (27/100) |
|
124 |
-
|
|
125 |
-
|
|
126 |
-
|
|
127 |
-
|
|
128 |
-
|
|
129 |
-
| astronomy | 0.2566 (39/152) |
|
130 |
-
| high_school_us_history | 0.2549 (52/204) |
|
131 |
| econometrics | 0.2544 (29/114) |
|
132 |
-
|
|
133 |
-
|
|
134 |
-
|
|
135 |
-
|
|
136 |
-
|
|
137 |
-
|
|
138 |
-
|
|
139 |
-
| high_school_physics | 0.2384 (36/151) |
|
140 |
-
| human_aging | 0.2377 (53/223) |
|
141 |
-
| nutrition | 0.2353 (72/306) |
|
142 |
-
| jurisprudence | 0.2315 (25/108) |
|
143 |
-
| business_ethics | 0.2300 (23/100) |
|
144 |
| high_school_government_and_politics | 0.2280 (44/193) |
|
145 |
-
|
|
146 |
-
|
|
147 |
-
|
|
|
|
148 |
| high_school_psychology | 0.2220 (121/545) |
|
149 |
-
|
|
150 |
-
|
|
|
|
|
|
151 |
| human_sexuality | 0.2137 (28/131) |
|
152 |
-
|
|
153 |
-
|
|
154 |
-
|
|
155 |
| machine_learning | 0.2054 (23/112) |
|
156 |
-
| high_school_macroeconomics | 0.2051 (80/390) |
|
157 |
-
| virology | 0.2048 (34/166) |
|
158 |
| management | 0.1845 (19/103) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
159 |
| professional_medicine | 0.1654 (45/272) |
|
160 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
161 |
|
162 |
---
|
|
|
33 |
|
34 |
## Training Data 📚
|
35 |
|
36 |
+
This model was trained on the **same dataset** used for Argonne-1.0 pretraining. Total processed tokens: 15,453,927,424.
|
37 |
|
38 |
---
|
39 |
|
|
|
98 |
## **Overall Accuracy**
|
99 |
**0.2549 (3579/14042)**
|
100 |
|
101 |
+
## Breakdown by Subject (Alphabetical Order):
|
102 |
| Subject | Accuracy (Correct/Total) |
|
103 |
|------------------------------|----------------------|
|
|
|
|
|
104 |
| abstract_algebra | 0.3200 (32/100) |
|
105 |
+
| anatomy | 0.3333 (45/135) |
|
106 |
+
| astronomy | 0.2566 (39/152) |
|
107 |
+
| business_ethics | 0.2300 (23/100) |
|
108 |
+
| clinical_knowledge | 0.2226 (59/265) |
|
109 |
| college_biology | 0.3056 (44/144) |
|
110 |
+
| college_chemistry | 0.2100 (21/100) |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
111 |
| college_computer_science | 0.2700 (27/100) |
|
112 |
+
| college_medicine | 0.2543 (44/173) |
|
113 |
| college_mathematics | 0.2700 (27/100) |
|
114 |
+
| college_physics | 0.2255 (23/102) |
|
115 |
+
| computer_security | 0.2900 (29/100) |
|
116 |
+
| conceptual_physics | 0.2213 (52/235) |
|
117 |
+
| electrical_engineering | 0.2759 (40/145) |
|
118 |
+
| elementary_mathematics | 0.2963 (112/378) |
|
|
|
|
|
119 |
| econometrics | 0.2544 (29/114) |
|
120 |
+
| formal_logic | 0.1508 (19/126) |
|
121 |
+
| global_facts | 0.3100 (31/100) |
|
122 |
+
| high_school_biology | 0.2613 (81/310) |
|
123 |
+
| high_school_chemistry | 0.3054 (62/203) |
|
124 |
+
| high_school_computer_science | 0.3100 (31/100) |
|
125 |
+
| high_school_european_history | 0.2970 (49/165) |
|
126 |
+
| high_school_geography | 0.2626 (52/198) |
|
|
|
|
|
|
|
|
|
|
|
127 |
| high_school_government_and_politics | 0.2280 (44/193) |
|
128 |
+
| high_school_macroeconomics | 0.2051 (80/390) |
|
129 |
+
| high_school_mathematics | 0.2630 (71/270) |
|
130 |
+
| high_school_microeconomics | 0.2059 (49/238) |
|
131 |
+
| high_school_physics | 0.2384 (36/151) |
|
132 |
| high_school_psychology | 0.2220 (121/545) |
|
133 |
+
| high_school_statistics | 0.2222 (48/216) |
|
134 |
+
| high_school_us_history | 0.2549 (52/204) |
|
135 |
+
| high_school_world_history | 0.2658 (63/237) |
|
136 |
+
| human_aging | 0.2377 (53/223) |
|
137 |
| human_sexuality | 0.2137 (28/131) |
|
138 |
+
| international_law | 0.3636 (44/121) |
|
139 |
+
| jurisprudence | 0.2315 (25/108) |
|
140 |
+
| logical_fallacies | 0.2945 (48/163) |
|
141 |
| machine_learning | 0.2054 (23/112) |
|
|
|
|
|
142 |
| management | 0.1845 (19/103) |
|
143 |
+
| marketing | 0.2436 (57/234) |
|
144 |
+
| medical_genetics | 0.2100 (21/100) |
|
145 |
+
| miscellaneous | 0.2439 (191/783) |
|
146 |
+
| moral_disputes | 0.2803 (97/346) |
|
147 |
+
| moral_scenarios | 0.2469 (221/895) |
|
148 |
+
| nutrition | 0.2353 (72/306) |
|
149 |
+
| philosophy | 0.3055 (95/311) |
|
150 |
+
| prehistory | 0.3025 (98/324) |
|
151 |
+
| professional_accounting | 0.2766 (78/282) |
|
152 |
+
| professional_law | 0.2692 (413/1534) |
|
153 |
| professional_medicine | 0.1654 (45/272) |
|
154 |
+
| professional_psychology | 0.2827 (173/612) |
|
155 |
+
| public_relations | 0.2182 (24/110) |
|
156 |
+
| security_studies | 0.2449 (60/245) |
|
157 |
+
| sociology | 0.2388 (48/201) |
|
158 |
+
| us_foreign_policy | 0.2500 (25/100) |
|
159 |
+
| virology | 0.2048 (34/166) |
|
160 |
+
| world_religions | 0.3041 (52/171) |
|
161 |
|
162 |
---
|