results
Browse files
all_results.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_HasAns_exact": 69.65206535309125,
|
4 |
+
"eval_HasAns_f1": 76.1771308342428,
|
5 |
+
"eval_HasAns_total": 18423,
|
6 |
+
"eval_NoAns_exact": 77.6604520706154,
|
7 |
+
"eval_NoAns_f1": 77.6604520706154,
|
8 |
+
"eval_NoAns_total": 6061,
|
9 |
+
"eval_best_exact": 71.63453684038556,
|
10 |
+
"eval_best_exact_thresh": 0.0,
|
11 |
+
"eval_best_f1": 76.5443261460238,
|
12 |
+
"eval_best_f1_thresh": 0.0,
|
13 |
+
"eval_exact": 71.63453684038556,
|
14 |
+
"eval_f1": 76.54432614602389,
|
15 |
+
"eval_samples": 44895,
|
16 |
+
"eval_total": 24484,
|
17 |
+
"train_loss": 0.8973765432098766,
|
18 |
+
"train_runtime": 1588.8114,
|
19 |
+
"train_samples": 317774,
|
20 |
+
"train_samples_per_second": 614.369,
|
21 |
+
"train_steps_per_second": 2.401
|
22 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_HasAns_exact": 69.65206535309125,
|
4 |
+
"eval_HasAns_f1": 76.1771308342428,
|
5 |
+
"eval_HasAns_total": 18423,
|
6 |
+
"eval_NoAns_exact": 77.6604520706154,
|
7 |
+
"eval_NoAns_f1": 77.6604520706154,
|
8 |
+
"eval_NoAns_total": 6061,
|
9 |
+
"eval_best_exact": 71.63453684038556,
|
10 |
+
"eval_best_exact_thresh": 0.0,
|
11 |
+
"eval_best_f1": 76.5443261460238,
|
12 |
+
"eval_best_f1_thresh": 0.0,
|
13 |
+
"eval_exact": 71.63453684038556,
|
14 |
+
"eval_f1": 76.54432614602389,
|
15 |
+
"eval_samples": 44895,
|
16 |
+
"eval_total": 24484
|
17 |
+
}
|
runs/Aug17_03-27-54_ip-172-31-85-26/events.out.tfevents.1660708828.ip-172-31-85-26.11771.2
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:67eecd5823bf8d520ff773029b45530f3a431f4de68e0d16408adc543ec6b01a
|
3 |
+
size 785
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"train_loss": 0.8973765432098766,
|
4 |
+
"train_runtime": 1588.8114,
|
5 |
+
"train_samples": 317774,
|
6 |
+
"train_samples_per_second": 614.369,
|
7 |
+
"train_steps_per_second": 2.401
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,298 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0,
|
5 |
+
"global_step": 3726,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.08,
|
12 |
+
"learning_rate": 6.81213097155126e-05,
|
13 |
+
"loss": 2.26,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.16,
|
18 |
+
"learning_rate": 6.624261943102522e-05,
|
19 |
+
"loss": 1.42,
|
20 |
+
"step": 200
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.24,
|
24 |
+
"learning_rate": 6.436392914653783e-05,
|
25 |
+
"loss": 1.28,
|
26 |
+
"step": 300
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.32,
|
30 |
+
"learning_rate": 6.248523886205046e-05,
|
31 |
+
"loss": 1.215,
|
32 |
+
"step": 400
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.4,
|
36 |
+
"learning_rate": 6.0606548577563064e-05,
|
37 |
+
"loss": 1.15,
|
38 |
+
"step": 500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.48,
|
42 |
+
"learning_rate": 5.872785829307568e-05,
|
43 |
+
"loss": 1.085,
|
44 |
+
"step": 600
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.56,
|
48 |
+
"learning_rate": 5.6849168008588294e-05,
|
49 |
+
"loss": 1.09,
|
50 |
+
"step": 700
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.64,
|
54 |
+
"learning_rate": 5.497047772410091e-05,
|
55 |
+
"loss": 1.07,
|
56 |
+
"step": 800
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.72,
|
60 |
+
"learning_rate": 5.3091787439613525e-05,
|
61 |
+
"loss": 1.035,
|
62 |
+
"step": 900
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.81,
|
66 |
+
"learning_rate": 5.121309715512614e-05,
|
67 |
+
"loss": 1.03,
|
68 |
+
"step": 1000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.89,
|
72 |
+
"learning_rate": 4.933440687063875e-05,
|
73 |
+
"loss": 0.99,
|
74 |
+
"step": 1100
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.97,
|
78 |
+
"learning_rate": 4.7455716586151364e-05,
|
79 |
+
"loss": 0.97,
|
80 |
+
"step": 1200
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 1.0,
|
84 |
+
"eval_HasAns_exact": 67.12804646365956,
|
85 |
+
"eval_HasAns_f1": 73.77021174615072,
|
86 |
+
"eval_HasAns_total": 18423,
|
87 |
+
"eval_NoAns_exact": 71.52285101468405,
|
88 |
+
"eval_NoAns_f1": 71.52285101468405,
|
89 |
+
"eval_NoAns_total": 6061,
|
90 |
+
"eval_best_exact": 68.21597778140827,
|
91 |
+
"eval_best_exact_thresh": 0.0,
|
92 |
+
"eval_best_f1": 73.21387890047849,
|
93 |
+
"eval_best_f1_thresh": 0.0,
|
94 |
+
"eval_exact": 68.21597778140827,
|
95 |
+
"eval_f1": 73.21387890047855,
|
96 |
+
"eval_total": 24484,
|
97 |
+
"step": 1242
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 1.05,
|
101 |
+
"learning_rate": 4.557702630166398e-05,
|
102 |
+
"loss": 0.895,
|
103 |
+
"step": 1300
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 1.13,
|
107 |
+
"learning_rate": 4.3698336017176595e-05,
|
108 |
+
"loss": 0.83,
|
109 |
+
"step": 1400
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 1.21,
|
113 |
+
"learning_rate": 4.18196457326892e-05,
|
114 |
+
"loss": 0.845,
|
115 |
+
"step": 1500
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 1.29,
|
119 |
+
"learning_rate": 3.994095544820182e-05,
|
120 |
+
"loss": 0.845,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 1.37,
|
125 |
+
"learning_rate": 3.8062265163714434e-05,
|
126 |
+
"loss": 0.84,
|
127 |
+
"step": 1700
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 1.45,
|
131 |
+
"learning_rate": 3.618357487922705e-05,
|
132 |
+
"loss": 0.825,
|
133 |
+
"step": 1800
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 1.53,
|
137 |
+
"learning_rate": 3.4304884594739665e-05,
|
138 |
+
"loss": 0.835,
|
139 |
+
"step": 1900
|
140 |
+
},
|
141 |
+
{
|
142 |
+
"epoch": 1.61,
|
143 |
+
"learning_rate": 3.242619431025228e-05,
|
144 |
+
"loss": 0.825,
|
145 |
+
"step": 2000
|
146 |
+
},
|
147 |
+
{
|
148 |
+
"epoch": 1.69,
|
149 |
+
"learning_rate": 3.0547504025764895e-05,
|
150 |
+
"loss": 0.825,
|
151 |
+
"step": 2100
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 1.77,
|
155 |
+
"learning_rate": 2.8668813741277507e-05,
|
156 |
+
"loss": 0.795,
|
157 |
+
"step": 2200
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 1.85,
|
161 |
+
"learning_rate": 2.679012345679012e-05,
|
162 |
+
"loss": 0.8,
|
163 |
+
"step": 2300
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 1.93,
|
167 |
+
"learning_rate": 2.4911433172302738e-05,
|
168 |
+
"loss": 0.795,
|
169 |
+
"step": 2400
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 2.0,
|
173 |
+
"eval_HasAns_exact": 68.19736199316073,
|
174 |
+
"eval_HasAns_f1": 74.61586025418268,
|
175 |
+
"eval_HasAns_total": 18423,
|
176 |
+
"eval_NoAns_exact": 76.17554858934169,
|
177 |
+
"eval_NoAns_f1": 76.17554858934169,
|
178 |
+
"eval_NoAns_total": 6061,
|
179 |
+
"eval_best_exact": 70.17235745793171,
|
180 |
+
"eval_best_exact_thresh": 0.0,
|
181 |
+
"eval_best_f1": 75.00196019697727,
|
182 |
+
"eval_best_f1_thresh": 0.0,
|
183 |
+
"eval_exact": 70.17235745793171,
|
184 |
+
"eval_f1": 75.00196019697732,
|
185 |
+
"eval_total": 24484,
|
186 |
+
"step": 2484
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 2.01,
|
190 |
+
"learning_rate": 2.303274288781535e-05,
|
191 |
+
"loss": 0.795,
|
192 |
+
"step": 2500
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 2.09,
|
196 |
+
"learning_rate": 2.1154052603327965e-05,
|
197 |
+
"loss": 0.67,
|
198 |
+
"step": 2600
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 2.17,
|
202 |
+
"learning_rate": 1.9275362318840577e-05,
|
203 |
+
"loss": 0.665,
|
204 |
+
"step": 2700
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 2.25,
|
208 |
+
"learning_rate": 1.7396672034353192e-05,
|
209 |
+
"loss": 0.68,
|
210 |
+
"step": 2800
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 2.33,
|
214 |
+
"learning_rate": 1.5517981749865808e-05,
|
215 |
+
"loss": 0.675,
|
216 |
+
"step": 2900
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 2.42,
|
220 |
+
"learning_rate": 1.3639291465378421e-05,
|
221 |
+
"loss": 0.665,
|
222 |
+
"step": 3000
|
223 |
+
},
|
224 |
+
{
|
225 |
+
"epoch": 2.5,
|
226 |
+
"learning_rate": 1.1760601180891035e-05,
|
227 |
+
"loss": 0.655,
|
228 |
+
"step": 3100
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 2.58,
|
232 |
+
"learning_rate": 9.881910896403648e-06,
|
233 |
+
"loss": 0.66,
|
234 |
+
"step": 3200
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 2.66,
|
238 |
+
"learning_rate": 8.003220611916264e-06,
|
239 |
+
"loss": 0.65,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 2.74,
|
244 |
+
"learning_rate": 6.124530327428877e-06,
|
245 |
+
"loss": 0.65,
|
246 |
+
"step": 3400
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 2.82,
|
250 |
+
"learning_rate": 4.245840042941492e-06,
|
251 |
+
"loss": 0.65,
|
252 |
+
"step": 3500
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 2.9,
|
256 |
+
"learning_rate": 2.367149758454106e-06,
|
257 |
+
"loss": 0.65,
|
258 |
+
"step": 3600
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 2.98,
|
262 |
+
"learning_rate": 4.884594739667203e-07,
|
263 |
+
"loss": 0.645,
|
264 |
+
"step": 3700
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 3.0,
|
268 |
+
"eval_HasAns_exact": 69.65206535309125,
|
269 |
+
"eval_HasAns_f1": 76.1771308342428,
|
270 |
+
"eval_HasAns_total": 18423,
|
271 |
+
"eval_NoAns_exact": 77.6604520706154,
|
272 |
+
"eval_NoAns_f1": 77.6604520706154,
|
273 |
+
"eval_NoAns_total": 6061,
|
274 |
+
"eval_best_exact": 71.63453684038556,
|
275 |
+
"eval_best_exact_thresh": 0.0,
|
276 |
+
"eval_best_f1": 76.5443261460238,
|
277 |
+
"eval_best_f1_thresh": 0.0,
|
278 |
+
"eval_exact": 71.63453684038556,
|
279 |
+
"eval_f1": 76.54432614602389,
|
280 |
+
"eval_total": 24484,
|
281 |
+
"step": 3726
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 3.0,
|
285 |
+
"step": 3726,
|
286 |
+
"total_flos": 1.8682613172115866e+17,
|
287 |
+
"train_loss": 0.8973765432098766,
|
288 |
+
"train_runtime": 1588.8114,
|
289 |
+
"train_samples_per_second": 614.369,
|
290 |
+
"train_steps_per_second": 2.401
|
291 |
+
}
|
292 |
+
],
|
293 |
+
"max_steps": 3726,
|
294 |
+
"num_train_epochs": 3,
|
295 |
+
"total_flos": 1.8682613172115866e+17,
|
296 |
+
"trial_name": null,
|
297 |
+
"trial_params": null
|
298 |
+
}
|