STS average to STS
Browse files
app.py
CHANGED
@@ -34,7 +34,7 @@ results = [
|
|
34 |
'Model Size (Million Parameters)': 279,
|
35 |
'Embedding Dimensions': 768,
|
36 |
'Average (8 datasets)': 37.95,
|
37 |
-
'STS
|
38 |
'Classification (3 datasets)': 58.42,
|
39 |
'PairClassification (1 datasets)': 57.62,
|
40 |
'Retrieval (3 datasets)': 5.57,
|
@@ -44,7 +44,7 @@ results = [
|
|
44 |
'Model Size (Million Parameters)': 561,
|
45 |
'Embedding Dimensions': 1024,
|
46 |
'Average (8 datasets)': 38.59,
|
47 |
-
'STS
|
48 |
'Classification (3 datasets)': 59.51,
|
49 |
'PairClassification (1 datasets)': 54.56,
|
50 |
'Retrieval (3 datasets)': 11.80,
|
@@ -54,7 +54,7 @@ results = [
|
|
54 |
'Model Size (Million Parameters)': 106,
|
55 |
'Embedding Dimensions': 768,
|
56 |
'Average (8 datasets)': 36.34,
|
57 |
-
'STS
|
58 |
'Classification (3 datasets)': 55.46,
|
59 |
'PairClassification (1 datasets)': 52.96,
|
60 |
'Retrieval (3 datasets)': 19.49,
|
@@ -64,7 +64,7 @@ results = [
|
|
64 |
'Model Size (Million Parameters)': 278,
|
65 |
'Embedding Dimensions': 768,
|
66 |
'Average (8 datasets)': 55.38,
|
67 |
-
'STS
|
68 |
'Classification (3 datasets)': 59.90,
|
69 |
'PairClassification (1 datasets)': 59.67,
|
70 |
'Retrieval (3 datasets)': 56.31,
|
@@ -74,7 +74,7 @@ results = [
|
|
74 |
'Model Size (Million Parameters)': 278,
|
75 |
'Embedding Dimensions': 768,
|
76 |
'Average (8 datasets)': 66.14,
|
77 |
-
'STS
|
78 |
'Classification (3 datasets)': 56.89,
|
79 |
'PairClassification (1 datasets)': 84.14,
|
80 |
'Retrieval (3 datasets)': 64.13,
|
@@ -84,7 +84,7 @@ results = [
|
|
84 |
'Model Size (Million Parameters)': 135,
|
85 |
'Embedding Dimensions': 512,
|
86 |
'Average (8 datasets)': 51.45,
|
87 |
-
'STS
|
88 |
'Classification (3 datasets)': 50.93,
|
89 |
'PairClassification (1 datasets)': 65.94,
|
90 |
'Retrieval (3 datasets)': 42.72,
|
@@ -94,7 +94,7 @@ results = [
|
|
94 |
'Model Size (Million Parameters)': 570,
|
95 |
'Embedding Dimensions': 1024,
|
96 |
'Average (8 datasets)': 75.64,
|
97 |
-
'STS
|
98 |
'Classification (3 datasets)': 59.95,
|
99 |
'PairClassification (1 datasets)': 79.02,
|
100 |
'Retrieval (3 datasets)': 91.42,
|
@@ -104,7 +104,7 @@ results = [
|
|
104 |
'Model Size (Million Parameters)': 279,
|
105 |
'Embedding Dimensions': 768,
|
106 |
'Average (8 datasets)': 53.83,
|
107 |
-
'STS
|
108 |
'Classification (3 datasets)': 49.44,
|
109 |
'PairClassification (1 datasets)': 61.87,
|
110 |
'Retrieval (3 datasets)': 54.17,
|
@@ -114,7 +114,7 @@ results = [
|
|
114 |
'Model Size (Million Parameters)': 106,
|
115 |
'Embedding Dimensions': 768,
|
116 |
'Average (8 datasets)': 54.01,
|
117 |
-
'STS
|
118 |
'Classification (3 datasets)': 56.71,
|
119 |
'PairClassification (1 datasets)': 59.14,
|
120 |
'Retrieval (3 datasets)': 51.05,
|
@@ -124,7 +124,7 @@ results = [
|
|
124 |
'Model Size (Million Parameters)': 278,
|
125 |
'Embedding Dimensions': 768,
|
126 |
'Average (8 datasets)': 60.02,
|
127 |
-
'STS
|
128 |
'Classification (3 datasets)': 53.50,
|
129 |
'PairClassification (1 datasets)': 63.35,
|
130 |
'Retrieval (3 datasets)': 66.05,
|
@@ -134,7 +134,7 @@ results = [
|
|
134 |
'Model Size (Million Parameters)': 279,
|
135 |
'Embedding Dimensions': 768,
|
136 |
'Average (8 datasets)': 57.69,
|
137 |
-
'STS
|
138 |
'Classification (3 datasets)': 55.93,
|
139 |
'PairClassification (1 datasets)': 66.49,
|
140 |
'Retrieval (3 datasets)': 54.90,
|
@@ -144,7 +144,7 @@ results = [
|
|
144 |
'Model Size (Million Parameters)': 106,
|
145 |
'Embedding Dimensions': 768,
|
146 |
'Average (8 datasets)': 62.22,
|
147 |
-
'STS
|
148 |
'Classification (3 datasets)': 59.19,
|
149 |
'PairClassification (1 datasets)': 67.04,
|
150 |
'Retrieval (3 datasets)': 63.83,
|
@@ -154,7 +154,7 @@ results = [
|
|
154 |
'Model Size (Million Parameters)': 278,
|
155 |
'Embedding Dimensions': 768,
|
156 |
'Average (8 datasets)': 63.28,
|
157 |
-
'STS
|
158 |
'Classification (3 datasets)': 58.77,
|
159 |
'PairClassification (1 datasets)': 65.87,
|
160 |
'Retrieval (3 datasets)': 66.20,
|
@@ -164,7 +164,7 @@ results = [
|
|
164 |
'Model Size (Million Parameters)': 279,
|
165 |
'Embedding Dimensions': 768,
|
166 |
'Average (8 datasets)': 65.37,
|
167 |
-
'STS
|
168 |
'Classification (3 datasets)': 56.87,
|
169 |
'PairClassification (1 datasets)': 79.78,
|
170 |
'Retrieval (3 datasets)': 65.02,
|
@@ -174,7 +174,7 @@ results = [
|
|
174 |
'Model Size (Million Parameters)': 106,
|
175 |
'Embedding Dimensions': 768,
|
176 |
'Average (8 datasets)': 63.55,
|
177 |
-
'STS
|
178 |
'Classification (3 datasets)': 56.33,
|
179 |
'PairClassification (1 datasets)': 77.04,
|
180 |
'Retrieval (3 datasets)': 62.38,
|
@@ -184,7 +184,7 @@ results = [
|
|
184 |
'Model Size (Million Parameters)': 278,
|
185 |
'Embedding Dimensions': 768,
|
186 |
'Average (8 datasets)': 66.00,
|
187 |
-
'STS
|
188 |
'Classification (3 datasets)': 57.27,
|
189 |
'PairClassification (1 datasets)': 77.84,
|
190 |
'Retrieval (3 datasets)': 67.94,
|
@@ -194,7 +194,7 @@ results = [
|
|
194 |
'Model Size (Million Parameters)': 279,
|
195 |
'Embedding Dimensions': 768,
|
196 |
'Average (8 datasets)': 66.84,
|
197 |
-
'STS
|
198 |
'Classification (3 datasets)': 56.90,
|
199 |
'PairClassification (1 datasets)': 81.47,
|
200 |
'Retrieval (3 datasets)': 68.03,
|
@@ -204,7 +204,7 @@ results = [
|
|
204 |
'Model Size (Million Parameters)': 106,
|
205 |
'Embedding Dimensions': 768,
|
206 |
'Average (8 datasets)': 67.17,
|
207 |
-
'STS
|
208 |
'Classification (3 datasets)': 58.16,
|
209 |
'PairClassification (1 datasets)': 82.43,
|
210 |
'Retrieval (3 datasets)': 67.66,
|
@@ -214,7 +214,7 @@ results = [
|
|
214 |
'Model Size (Million Parameters)': 278,
|
215 |
'Embedding Dimensions': 768,
|
216 |
'Average (8 datasets)': 66.94,
|
217 |
-
'STS
|
218 |
'Classification (3 datasets)': 57.63,
|
219 |
'PairClassification (1 datasets)': 81.01,
|
220 |
'Retrieval (3 datasets)': 68.04,
|
@@ -224,7 +224,7 @@ results = [
|
|
224 |
'Model Size (Million Parameters)': 7110,
|
225 |
'Embedding Dimensions': 4096,
|
226 |
'Average (8 datasets)': 71.94,
|
227 |
-
'STS
|
228 |
'Classification (3 datasets)': 60.46,
|
229 |
'PairClassification (1 datasets)': 68.04,
|
230 |
'Retrieval (3 datasets)': 86.80,
|
@@ -234,7 +234,7 @@ results = [
|
|
234 |
'Model Size (Million Parameters)': 7610,
|
235 |
'Embedding Dimensions': 3584,
|
236 |
'Average (8 datasets)': 49.31,
|
237 |
-
'STS
|
238 |
'Classification (3 datasets)': 57.55,
|
239 |
'PairClassification (1 datasets)': 61.73,
|
240 |
'Retrieval (3 datasets)': 38.31,
|
@@ -244,7 +244,7 @@ results = [
|
|
244 |
'Model Size (Million Parameters)': 7240,
|
245 |
'Embedding Dimensions': 4096,
|
246 |
'Average (8 datasets)': 42.38,
|
247 |
-
'STS
|
248 |
'Classification (3 datasets)': 56.83,
|
249 |
'PairClassification (1 datasets)': 56.40,
|
250 |
'Retrieval (3 datasets)': 22.79,
|
@@ -255,7 +255,7 @@ results = [
|
|
255 |
'Model Size (Million Parameters)': 8030,
|
256 |
'Embedding Dimensions': 4096,
|
257 |
'Average (8 datasets)': 51.63,
|
258 |
-
'STS
|
259 |
'Classification (3 datasets)': 58.54,
|
260 |
'PairClassification (1 datasets)': 57.76,
|
261 |
'Retrieval (3 datasets)': 47.93,
|
@@ -265,7 +265,7 @@ results = [
|
|
265 |
'Model Size (Million Parameters)': 8030,
|
266 |
'Embedding Dimensions': 4096,
|
267 |
'Average (8 datasets)': 52.81,
|
268 |
-
'STS
|
269 |
'Classification (3 datasets)': 58.85,
|
270 |
'PairClassification (1 datasets)': 58.04,
|
271 |
'Retrieval (3 datasets)': 50.38,
|
@@ -275,7 +275,7 @@ results = [
|
|
275 |
'Model Size (Million Parameters)': 8030,
|
276 |
'Embedding Dimensions': 4096,
|
277 |
'Average (8 datasets)': 50.36,
|
278 |
-
'STS
|
279 |
'Classification (3 datasets)': 58.18,
|
280 |
'PairClassification (1 datasets)': 58.12,
|
281 |
'Retrieval (3 datasets)': 43.64,
|
@@ -285,7 +285,7 @@ results = [
|
|
285 |
'Model Size (Million Parameters)': 8030,
|
286 |
'Embedding Dimensions': 4096,
|
287 |
'Average (8 datasets)': 50.06,
|
288 |
-
'STS
|
289 |
'Classification (3 datasets)': 57.90,
|
290 |
'PairClassification (1 datasets)': 57.47,
|
291 |
'Retrieval (3 datasets)': 43.63,
|
@@ -295,7 +295,7 @@ results = [
|
|
295 |
'Model Size (Million Parameters)': 8030,
|
296 |
'Embedding Dimensions': 4096,
|
297 |
'Average (8 datasets)': 53.51,
|
298 |
-
'STS
|
299 |
'Classification (3 datasets)': 58.91,
|
300 |
'PairClassification (1 datasets)': 58.05,
|
301 |
'Retrieval (3 datasets)': 52.65,
|
@@ -305,7 +305,7 @@ results = [
|
|
305 |
'Model Size (Million Parameters)': "N/A",
|
306 |
'Embedding Dimensions': 768,
|
307 |
'Average (8 datasets)': 68.01,
|
308 |
-
'STS
|
309 |
'Classification (3 datasets)': 57.31,
|
310 |
'PairClassification (1 datasets)': 62.03,
|
311 |
'Retrieval (3 datasets)': 85.23,
|
@@ -315,7 +315,7 @@ results = [
|
|
315 |
'Model Size (Million Parameters)': "N/A",
|
316 |
'Embedding Dimensions': 1024,
|
317 |
'Average (8 datasets)': 74.86,
|
318 |
-
'STS
|
319 |
'Classification (3 datasets)': 59.96,
|
320 |
'PairClassification (1 datasets)': 73.28,
|
321 |
'Retrieval (3 datasets)': 91.43,
|
@@ -325,7 +325,7 @@ results = [
|
|
325 |
'Model Size (Million Parameters)': "N/A",
|
326 |
'Embedding Dimensions': 3072,
|
327 |
'Average (8 datasets)': 69.26,
|
328 |
-
'STS
|
329 |
'Classification (3 datasets)': 58.79,
|
330 |
'PairClassification (1 datasets)': 67.33,
|
331 |
'Retrieval (3 datasets)': 83.87,
|
@@ -337,7 +337,7 @@ results = [
|
|
337 |
{
|
338 |
**result,
|
339 |
'Average (8 datasets)': round(sum(
|
340 |
-
result.get(key, 0) for key in ['STS
|
341 |
) / 4, 2),
|
342 |
}
|
343 |
for result in results
|
|
|
34 |
'Model Size (Million Parameters)': 279,
|
35 |
'Embedding Dimensions': 768,
|
36 |
'Average (8 datasets)': 37.95,
|
37 |
+
'STS (1 datasets)': 44.48,
|
38 |
'Classification (3 datasets)': 58.42,
|
39 |
'PairClassification (1 datasets)': 57.62,
|
40 |
'Retrieval (3 datasets)': 5.57,
|
|
|
44 |
'Model Size (Million Parameters)': 561,
|
45 |
'Embedding Dimensions': 1024,
|
46 |
'Average (8 datasets)': 38.59,
|
47 |
+
'STS (1 datasets)': 38.31,
|
48 |
'Classification (3 datasets)': 59.51,
|
49 |
'PairClassification (1 datasets)': 54.56,
|
50 |
'Retrieval (3 datasets)': 11.80,
|
|
|
54 |
'Model Size (Million Parameters)': 106,
|
55 |
'Embedding Dimensions': 768,
|
56 |
'Average (8 datasets)': 36.34,
|
57 |
+
'STS (1 datasets)': 21.32,
|
58 |
'Classification (3 datasets)': 55.46,
|
59 |
'PairClassification (1 datasets)': 52.96,
|
60 |
'Retrieval (3 datasets)': 19.49,
|
|
|
64 |
'Model Size (Million Parameters)': 278,
|
65 |
'Embedding Dimensions': 768,
|
66 |
'Average (8 datasets)': 55.38,
|
67 |
+
'STS (1 datasets)': 51.56,
|
68 |
'Classification (3 datasets)': 59.90,
|
69 |
'PairClassification (1 datasets)': 59.67,
|
70 |
'Retrieval (3 datasets)': 56.31,
|
|
|
74 |
'Model Size (Million Parameters)': 278,
|
75 |
'Embedding Dimensions': 768,
|
76 |
'Average (8 datasets)': 66.14,
|
77 |
+
'STS (1 datasets)': 80.49,
|
78 |
'Classification (3 datasets)': 56.89,
|
79 |
'PairClassification (1 datasets)': 84.14,
|
80 |
'Retrieval (3 datasets)': 64.13,
|
|
|
84 |
'Model Size (Million Parameters)': 135,
|
85 |
'Embedding Dimensions': 512,
|
86 |
'Average (8 datasets)': 51.45,
|
87 |
+
'STS (1 datasets)': 65.37,
|
88 |
'Classification (3 datasets)': 50.93,
|
89 |
'PairClassification (1 datasets)': 65.94,
|
90 |
'Retrieval (3 datasets)': 42.72,
|
|
|
94 |
'Model Size (Million Parameters)': 570,
|
95 |
'Embedding Dimensions': 1024,
|
96 |
'Average (8 datasets)': 75.64,
|
97 |
+
'STS (1 datasets)': 77.22,
|
98 |
'Classification (3 datasets)': 59.95,
|
99 |
'PairClassification (1 datasets)': 79.02,
|
100 |
'Retrieval (3 datasets)': 91.42,
|
|
|
104 |
'Model Size (Million Parameters)': 279,
|
105 |
'Embedding Dimensions': 768,
|
106 |
'Average (8 datasets)': 53.83,
|
107 |
+
'STS (1 datasets)': 63.98,
|
108 |
'Classification (3 datasets)': 49.44,
|
109 |
'PairClassification (1 datasets)': 61.87,
|
110 |
'Retrieval (3 datasets)': 54.17,
|
|
|
114 |
'Model Size (Million Parameters)': 106,
|
115 |
'Embedding Dimensions': 768,
|
116 |
'Average (8 datasets)': 54.01,
|
117 |
+
'STS (1 datasets)': 60.73,
|
118 |
'Classification (3 datasets)': 56.71,
|
119 |
'PairClassification (1 datasets)': 59.14,
|
120 |
'Retrieval (3 datasets)': 51.05,
|
|
|
124 |
'Model Size (Million Parameters)': 278,
|
125 |
'Embedding Dimensions': 768,
|
126 |
'Average (8 datasets)': 60.02,
|
127 |
+
'STS (1 datasets)': 67.82,
|
128 |
'Classification (3 datasets)': 53.50,
|
129 |
'PairClassification (1 datasets)': 63.35,
|
130 |
'Retrieval (3 datasets)': 66.05,
|
|
|
134 |
'Model Size (Million Parameters)': 279,
|
135 |
'Embedding Dimensions': 768,
|
136 |
'Average (8 datasets)': 57.69,
|
137 |
+
'STS (1 datasets)': 68.91,
|
138 |
'Classification (3 datasets)': 55.93,
|
139 |
'PairClassification (1 datasets)': 66.49,
|
140 |
'Retrieval (3 datasets)': 54.90,
|
|
|
144 |
'Model Size (Million Parameters)': 106,
|
145 |
'Embedding Dimensions': 768,
|
146 |
'Average (8 datasets)': 62.22,
|
147 |
+
'STS (1 datasets)': 71.35,
|
148 |
'Classification (3 datasets)': 59.19,
|
149 |
'PairClassification (1 datasets)': 67.04,
|
150 |
'Retrieval (3 datasets)': 63.83,
|
|
|
154 |
'Model Size (Million Parameters)': 278,
|
155 |
'Embedding Dimensions': 768,
|
156 |
'Average (8 datasets)': 63.28,
|
157 |
+
'STS (1 datasets)': 74.08,
|
158 |
'Classification (3 datasets)': 58.77,
|
159 |
'PairClassification (1 datasets)': 65.87,
|
160 |
'Retrieval (3 datasets)': 66.20,
|
|
|
164 |
'Model Size (Million Parameters)': 279,
|
165 |
'Embedding Dimensions': 768,
|
166 |
'Average (8 datasets)': 65.37,
|
167 |
+
'STS (1 datasets)': 78.78,
|
168 |
'Classification (3 datasets)': 56.87,
|
169 |
'PairClassification (1 datasets)': 79.78,
|
170 |
'Retrieval (3 datasets)': 65.02,
|
|
|
174 |
'Model Size (Million Parameters)': 106,
|
175 |
'Embedding Dimensions': 768,
|
176 |
'Average (8 datasets)': 63.55,
|
177 |
+
'STS (1 datasets)': 77.77,
|
178 |
'Classification (3 datasets)': 56.33,
|
179 |
'PairClassification (1 datasets)': 77.04,
|
180 |
'Retrieval (3 datasets)': 62.38,
|
|
|
184 |
'Model Size (Million Parameters)': 278,
|
185 |
'Embedding Dimensions': 768,
|
186 |
'Average (8 datasets)': 66.00,
|
187 |
+
'STS (1 datasets)': 77.80,
|
188 |
'Classification (3 datasets)': 57.27,
|
189 |
'PairClassification (1 datasets)': 77.84,
|
190 |
'Retrieval (3 datasets)': 67.94,
|
|
|
194 |
'Model Size (Million Parameters)': 279,
|
195 |
'Embedding Dimensions': 768,
|
196 |
'Average (8 datasets)': 66.84,
|
197 |
+
'STS (1 datasets)': 79.69,
|
198 |
'Classification (3 datasets)': 56.90,
|
199 |
'PairClassification (1 datasets)': 81.47,
|
200 |
'Retrieval (3 datasets)': 68.03,
|
|
|
204 |
'Model Size (Million Parameters)': 106,
|
205 |
'Embedding Dimensions': 768,
|
206 |
'Average (8 datasets)': 67.17,
|
207 |
+
'STS (1 datasets)': 78.78,
|
208 |
'Classification (3 datasets)': 58.16,
|
209 |
'PairClassification (1 datasets)': 82.43,
|
210 |
'Retrieval (3 datasets)': 67.66,
|
|
|
214 |
'Model Size (Million Parameters)': 278,
|
215 |
'Embedding Dimensions': 768,
|
216 |
'Average (8 datasets)': 66.94,
|
217 |
+
'STS (1 datasets)': 78.90,
|
218 |
'Classification (3 datasets)': 57.63,
|
219 |
'PairClassification (1 datasets)': 81.01,
|
220 |
'Retrieval (3 datasets)': 68.04,
|
|
|
224 |
'Model Size (Million Parameters)': 7110,
|
225 |
'Embedding Dimensions': 4096,
|
226 |
'Average (8 datasets)': 71.94,
|
227 |
+
'STS (1 datasets)': 75.52,
|
228 |
'Classification (3 datasets)': 60.46,
|
229 |
'PairClassification (1 datasets)': 68.04,
|
230 |
'Retrieval (3 datasets)': 86.80,
|
|
|
234 |
'Model Size (Million Parameters)': 7610,
|
235 |
'Embedding Dimensions': 3584,
|
236 |
'Average (8 datasets)': 49.31,
|
237 |
+
'STS (1 datasets)': 51.60,
|
238 |
'Classification (3 datasets)': 57.55,
|
239 |
'PairClassification (1 datasets)': 61.73,
|
240 |
'Retrieval (3 datasets)': 38.31,
|
|
|
244 |
'Model Size (Million Parameters)': 7240,
|
245 |
'Embedding Dimensions': 4096,
|
246 |
'Average (8 datasets)': 42.38,
|
247 |
+
'STS (1 datasets)': 45.50,
|
248 |
'Classification (3 datasets)': 56.83,
|
249 |
'PairClassification (1 datasets)': 56.40,
|
250 |
'Retrieval (3 datasets)': 22.79,
|
|
|
255 |
'Model Size (Million Parameters)': 8030,
|
256 |
'Embedding Dimensions': 4096,
|
257 |
'Average (8 datasets)': 51.63,
|
258 |
+
'STS (1 datasets)': 49.48,
|
259 |
'Classification (3 datasets)': 58.54,
|
260 |
'PairClassification (1 datasets)': 57.76,
|
261 |
'Retrieval (3 datasets)': 47.93,
|
|
|
265 |
'Model Size (Million Parameters)': 8030,
|
266 |
'Embedding Dimensions': 4096,
|
267 |
'Average (8 datasets)': 52.81,
|
268 |
+
'STS (1 datasets)': 50.63,
|
269 |
'Classification (3 datasets)': 58.85,
|
270 |
'PairClassification (1 datasets)': 58.04,
|
271 |
'Retrieval (3 datasets)': 50.38,
|
|
|
275 |
'Model Size (Million Parameters)': 8030,
|
276 |
'Embedding Dimensions': 4096,
|
277 |
'Average (8 datasets)': 50.36,
|
278 |
+
'STS (1 datasets)': 49.98,
|
279 |
'Classification (3 datasets)': 58.18,
|
280 |
'PairClassification (1 datasets)': 58.12,
|
281 |
'Retrieval (3 datasets)': 43.64,
|
|
|
285 |
'Model Size (Million Parameters)': 8030,
|
286 |
'Embedding Dimensions': 4096,
|
287 |
'Average (8 datasets)': 50.06,
|
288 |
+
'STS (1 datasets)': 49.76,
|
289 |
'Classification (3 datasets)': 57.90,
|
290 |
'PairClassification (1 datasets)': 57.47,
|
291 |
'Retrieval (3 datasets)': 43.63,
|
|
|
295 |
'Model Size (Million Parameters)': 8030,
|
296 |
'Embedding Dimensions': 4096,
|
297 |
'Average (8 datasets)': 53.51,
|
298 |
+
'STS (1 datasets)': 51.46,
|
299 |
'Classification (3 datasets)': 58.91,
|
300 |
'PairClassification (1 datasets)': 58.05,
|
301 |
'Retrieval (3 datasets)': 52.65,
|
|
|
305 |
'Model Size (Million Parameters)': "N/A",
|
306 |
'Embedding Dimensions': 768,
|
307 |
'Average (8 datasets)': 68.01,
|
308 |
+
'STS (1 datasets)': 68.03,
|
309 |
'Classification (3 datasets)': 57.31,
|
310 |
'PairClassification (1 datasets)': 62.03,
|
311 |
'Retrieval (3 datasets)': 85.23,
|
|
|
315 |
'Model Size (Million Parameters)': "N/A",
|
316 |
'Embedding Dimensions': 1024,
|
317 |
'Average (8 datasets)': 74.86,
|
318 |
+
'STS (1 datasets)': 77.87,
|
319 |
'Classification (3 datasets)': 59.96,
|
320 |
'PairClassification (1 datasets)': 73.28,
|
321 |
'Retrieval (3 datasets)': 91.43,
|
|
|
325 |
'Model Size (Million Parameters)': "N/A",
|
326 |
'Embedding Dimensions': 3072,
|
327 |
'Average (8 datasets)': 69.26,
|
328 |
+
'STS (1 datasets)': 70.46,
|
329 |
'Classification (3 datasets)': 58.79,
|
330 |
'PairClassification (1 datasets)': 67.33,
|
331 |
'Retrieval (3 datasets)': 83.87,
|
|
|
337 |
{
|
338 |
**result,
|
339 |
'Average (8 datasets)': round(sum(
|
340 |
+
result.get(key, 0) for key in ['STS (1 datasets)', 'Classification (3 datasets)', 'PairClassification (1 datasets)', 'Retrieval (3 datasets)']
|
341 |
) / 4, 2),
|
342 |
}
|
343 |
for result in results
|