tiagoblima
commited on
Commit
•
dc469a1
1
Parent(s):
bc6eec7
End of training
Browse files- README.md +3 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +776 -0
README.md
CHANGED
@@ -3,6 +3,8 @@ license: mit
|
|
3 |
base_model: facebook/mbart-large-50
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
|
|
6 |
model-index:
|
7 |
- name: mbart50-qg-ap-peft-oficial
|
8 |
results: []
|
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# mbart50-qg-ap-peft-oficial
|
15 |
|
16 |
-
This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- Loss: 2.5230
|
19 |
|
|
|
3 |
base_model: facebook/mbart-large-50
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- tiagoblima/du-qg-squadv1_pt
|
8 |
model-index:
|
9 |
- name: mbart50-qg-ap-peft-oficial
|
10 |
results: []
|
|
|
15 |
|
16 |
# mbart50-qg-ap-peft-oficial
|
17 |
|
18 |
+
This model is a fine-tuned version of [facebook/mbart-large-50](https://huggingface.co/facebook/mbart-large-50) on the tiagoblima/du-qg-squadv1_pt dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 2.5230
|
21 |
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 2.5229861736297607,
|
4 |
+
"eval_runtime": 57.5969,
|
5 |
+
"eval_samples": 10692,
|
6 |
+
"eval_samples_per_second": 185.635,
|
7 |
+
"eval_steps_per_second": 2.917,
|
8 |
+
"train_loss": 2.6363602522611114,
|
9 |
+
"train_runtime": 3204.9921,
|
10 |
+
"train_samples": 75722,
|
11 |
+
"train_samples_per_second": 118.131,
|
12 |
+
"train_steps_per_second": 3.693
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 2.5229861736297607,
|
4 |
+
"eval_runtime": 57.5969,
|
5 |
+
"eval_samples": 10692,
|
6 |
+
"eval_samples_per_second": 185.635,
|
7 |
+
"eval_steps_per_second": 2.917
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"train_loss": 2.6363602522611114,
|
4 |
+
"train_runtime": 3204.9921,
|
5 |
+
"train_samples": 75722,
|
6 |
+
"train_samples_per_second": 118.131,
|
7 |
+
"train_steps_per_second": 3.693
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,776 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 2.5229861736297607,
|
3 |
+
"best_model_checkpoint": "/temp/mbart50-qg-ap-peft-oficial/checkpoint-11835",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 11835,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.04,
|
13 |
+
"learning_rate": 9.915504858470639e-05,
|
14 |
+
"loss": 6.9931,
|
15 |
+
"step": 100
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.08,
|
19 |
+
"learning_rate": 9.831009716941276e-05,
|
20 |
+
"loss": 4.698,
|
21 |
+
"step": 200
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.13,
|
25 |
+
"learning_rate": 9.746514575411914e-05,
|
26 |
+
"loss": 3.4174,
|
27 |
+
"step": 300
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.17,
|
31 |
+
"learning_rate": 9.662019433882552e-05,
|
32 |
+
"loss": 3.1673,
|
33 |
+
"step": 400
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.21,
|
37 |
+
"learning_rate": 9.57752429235319e-05,
|
38 |
+
"loss": 3.1062,
|
39 |
+
"step": 500
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.25,
|
43 |
+
"learning_rate": 9.493029150823828e-05,
|
44 |
+
"loss": 3.1208,
|
45 |
+
"step": 600
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.3,
|
49 |
+
"learning_rate": 9.408534009294467e-05,
|
50 |
+
"loss": 3.0488,
|
51 |
+
"step": 700
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.34,
|
55 |
+
"learning_rate": 9.324038867765104e-05,
|
56 |
+
"loss": 3.0227,
|
57 |
+
"step": 800
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.38,
|
61 |
+
"learning_rate": 9.239543726235742e-05,
|
62 |
+
"loss": 2.9982,
|
63 |
+
"step": 900
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.42,
|
67 |
+
"learning_rate": 9.15504858470638e-05,
|
68 |
+
"loss": 2.9974,
|
69 |
+
"step": 1000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.46,
|
73 |
+
"learning_rate": 9.070553443177018e-05,
|
74 |
+
"loss": 2.9213,
|
75 |
+
"step": 1100
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.51,
|
79 |
+
"learning_rate": 8.986058301647655e-05,
|
80 |
+
"loss": 2.934,
|
81 |
+
"step": 1200
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.55,
|
85 |
+
"learning_rate": 8.901563160118294e-05,
|
86 |
+
"loss": 2.908,
|
87 |
+
"step": 1300
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.59,
|
91 |
+
"learning_rate": 8.817068018588932e-05,
|
92 |
+
"loss": 2.8976,
|
93 |
+
"step": 1400
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.63,
|
97 |
+
"learning_rate": 8.732572877059569e-05,
|
98 |
+
"loss": 2.9042,
|
99 |
+
"step": 1500
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.68,
|
103 |
+
"learning_rate": 8.648077735530208e-05,
|
104 |
+
"loss": 2.8954,
|
105 |
+
"step": 1600
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.72,
|
109 |
+
"learning_rate": 8.563582594000845e-05,
|
110 |
+
"loss": 2.8462,
|
111 |
+
"step": 1700
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.76,
|
115 |
+
"learning_rate": 8.479087452471483e-05,
|
116 |
+
"loss": 2.8818,
|
117 |
+
"step": 1800
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.8,
|
121 |
+
"learning_rate": 8.394592310942122e-05,
|
122 |
+
"loss": 2.8113,
|
123 |
+
"step": 1900
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.84,
|
127 |
+
"learning_rate": 8.31009716941276e-05,
|
128 |
+
"loss": 2.8679,
|
129 |
+
"step": 2000
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.89,
|
133 |
+
"learning_rate": 8.225602027883397e-05,
|
134 |
+
"loss": 2.7948,
|
135 |
+
"step": 2100
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.93,
|
139 |
+
"learning_rate": 8.141106886354036e-05,
|
140 |
+
"loss": 2.7912,
|
141 |
+
"step": 2200
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.97,
|
145 |
+
"learning_rate": 8.056611744824673e-05,
|
146 |
+
"loss": 2.8113,
|
147 |
+
"step": 2300
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 1.0,
|
151 |
+
"eval_loss": 2.6543636322021484,
|
152 |
+
"eval_runtime": 57.6249,
|
153 |
+
"eval_samples_per_second": 185.545,
|
154 |
+
"eval_steps_per_second": 2.915,
|
155 |
+
"step": 2367
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.01,
|
159 |
+
"learning_rate": 7.972116603295311e-05,
|
160 |
+
"loss": 2.7489,
|
161 |
+
"step": 2400
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.06,
|
165 |
+
"learning_rate": 7.887621461765948e-05,
|
166 |
+
"loss": 2.7036,
|
167 |
+
"step": 2500
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.1,
|
171 |
+
"learning_rate": 7.803126320236587e-05,
|
172 |
+
"loss": 2.7098,
|
173 |
+
"step": 2600
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 1.14,
|
177 |
+
"learning_rate": 7.718631178707225e-05,
|
178 |
+
"loss": 2.6703,
|
179 |
+
"step": 2700
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.18,
|
183 |
+
"learning_rate": 7.634136037177862e-05,
|
184 |
+
"loss": 2.6702,
|
185 |
+
"step": 2800
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 1.23,
|
189 |
+
"learning_rate": 7.5496408956485e-05,
|
190 |
+
"loss": 2.656,
|
191 |
+
"step": 2900
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 1.27,
|
195 |
+
"learning_rate": 7.465145754119138e-05,
|
196 |
+
"loss": 2.6548,
|
197 |
+
"step": 3000
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 1.31,
|
201 |
+
"learning_rate": 7.380650612589776e-05,
|
202 |
+
"loss": 2.6593,
|
203 |
+
"step": 3100
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.35,
|
207 |
+
"learning_rate": 7.296155471060415e-05,
|
208 |
+
"loss": 2.6228,
|
209 |
+
"step": 3200
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.39,
|
213 |
+
"learning_rate": 7.211660329531053e-05,
|
214 |
+
"loss": 2.662,
|
215 |
+
"step": 3300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 1.44,
|
219 |
+
"learning_rate": 7.12716518800169e-05,
|
220 |
+
"loss": 2.6567,
|
221 |
+
"step": 3400
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 1.48,
|
225 |
+
"learning_rate": 7.042670046472329e-05,
|
226 |
+
"loss": 2.6563,
|
227 |
+
"step": 3500
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 1.52,
|
231 |
+
"learning_rate": 6.958174904942966e-05,
|
232 |
+
"loss": 2.622,
|
233 |
+
"step": 3600
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 1.56,
|
237 |
+
"learning_rate": 6.873679763413604e-05,
|
238 |
+
"loss": 2.6196,
|
239 |
+
"step": 3700
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 1.61,
|
243 |
+
"learning_rate": 6.789184621884243e-05,
|
244 |
+
"loss": 2.642,
|
245 |
+
"step": 3800
|
246 |
+
},
|
247 |
+
{
|
248 |
+
"epoch": 1.65,
|
249 |
+
"learning_rate": 6.70468948035488e-05,
|
250 |
+
"loss": 2.6213,
|
251 |
+
"step": 3900
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 1.69,
|
255 |
+
"learning_rate": 6.620194338825518e-05,
|
256 |
+
"loss": 2.6472,
|
257 |
+
"step": 4000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 1.73,
|
261 |
+
"learning_rate": 6.535699197296155e-05,
|
262 |
+
"loss": 2.6091,
|
263 |
+
"step": 4100
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.77,
|
267 |
+
"learning_rate": 6.451204055766794e-05,
|
268 |
+
"loss": 2.6267,
|
269 |
+
"step": 4200
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.82,
|
273 |
+
"learning_rate": 6.366708914237431e-05,
|
274 |
+
"loss": 2.5979,
|
275 |
+
"step": 4300
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 1.86,
|
279 |
+
"learning_rate": 6.282213772708069e-05,
|
280 |
+
"loss": 2.6143,
|
281 |
+
"step": 4400
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 1.9,
|
285 |
+
"learning_rate": 6.197718631178708e-05,
|
286 |
+
"loss": 2.6189,
|
287 |
+
"step": 4500
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 1.94,
|
291 |
+
"learning_rate": 6.113223489649346e-05,
|
292 |
+
"loss": 2.6089,
|
293 |
+
"step": 4600
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 1.99,
|
297 |
+
"learning_rate": 6.028728348119983e-05,
|
298 |
+
"loss": 2.5839,
|
299 |
+
"step": 4700
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 2.0,
|
303 |
+
"eval_loss": 2.5805256366729736,
|
304 |
+
"eval_runtime": 57.5593,
|
305 |
+
"eval_samples_per_second": 185.756,
|
306 |
+
"eval_steps_per_second": 2.919,
|
307 |
+
"step": 4734
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 2.03,
|
311 |
+
"learning_rate": 5.944233206590622e-05,
|
312 |
+
"loss": 2.5342,
|
313 |
+
"step": 4800
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 2.07,
|
317 |
+
"learning_rate": 5.859738065061259e-05,
|
318 |
+
"loss": 2.5069,
|
319 |
+
"step": 4900
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 2.11,
|
323 |
+
"learning_rate": 5.775242923531897e-05,
|
324 |
+
"loss": 2.5306,
|
325 |
+
"step": 5000
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 2.15,
|
329 |
+
"learning_rate": 5.6907477820025357e-05,
|
330 |
+
"loss": 2.5291,
|
331 |
+
"step": 5100
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 2.2,
|
335 |
+
"learning_rate": 5.606252640473173e-05,
|
336 |
+
"loss": 2.5474,
|
337 |
+
"step": 5200
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 2.24,
|
341 |
+
"learning_rate": 5.521757498943811e-05,
|
342 |
+
"loss": 2.5059,
|
343 |
+
"step": 5300
|
344 |
+
},
|
345 |
+
{
|
346 |
+
"epoch": 2.28,
|
347 |
+
"learning_rate": 5.437262357414449e-05,
|
348 |
+
"loss": 2.5508,
|
349 |
+
"step": 5400
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 2.32,
|
353 |
+
"learning_rate": 5.3527672158850874e-05,
|
354 |
+
"loss": 2.5249,
|
355 |
+
"step": 5500
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"epoch": 2.37,
|
359 |
+
"learning_rate": 5.2682720743557245e-05,
|
360 |
+
"loss": 2.527,
|
361 |
+
"step": 5600
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 2.41,
|
365 |
+
"learning_rate": 5.183776932826363e-05,
|
366 |
+
"loss": 2.4997,
|
367 |
+
"step": 5700
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 2.45,
|
371 |
+
"learning_rate": 5.099281791297e-05,
|
372 |
+
"loss": 2.4881,
|
373 |
+
"step": 5800
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 2.49,
|
377 |
+
"learning_rate": 5.0147866497676385e-05,
|
378 |
+
"loss": 2.5018,
|
379 |
+
"step": 5900
|
380 |
+
},
|
381 |
+
{
|
382 |
+
"epoch": 2.53,
|
383 |
+
"learning_rate": 4.930291508238277e-05,
|
384 |
+
"loss": 2.4889,
|
385 |
+
"step": 6000
|
386 |
+
},
|
387 |
+
{
|
388 |
+
"epoch": 2.58,
|
389 |
+
"learning_rate": 4.845796366708915e-05,
|
390 |
+
"loss": 2.5156,
|
391 |
+
"step": 6100
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 2.62,
|
395 |
+
"learning_rate": 4.7613012251795525e-05,
|
396 |
+
"loss": 2.5002,
|
397 |
+
"step": 6200
|
398 |
+
},
|
399 |
+
{
|
400 |
+
"epoch": 2.66,
|
401 |
+
"learning_rate": 4.67680608365019e-05,
|
402 |
+
"loss": 2.501,
|
403 |
+
"step": 6300
|
404 |
+
},
|
405 |
+
{
|
406 |
+
"epoch": 2.7,
|
407 |
+
"learning_rate": 4.592310942120828e-05,
|
408 |
+
"loss": 2.5158,
|
409 |
+
"step": 6400
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 2.75,
|
413 |
+
"learning_rate": 4.507815800591466e-05,
|
414 |
+
"loss": 2.5346,
|
415 |
+
"step": 6500
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 2.79,
|
419 |
+
"learning_rate": 4.423320659062104e-05,
|
420 |
+
"loss": 2.4897,
|
421 |
+
"step": 6600
|
422 |
+
},
|
423 |
+
{
|
424 |
+
"epoch": 2.83,
|
425 |
+
"learning_rate": 4.338825517532742e-05,
|
426 |
+
"loss": 2.5013,
|
427 |
+
"step": 6700
|
428 |
+
},
|
429 |
+
{
|
430 |
+
"epoch": 2.87,
|
431 |
+
"learning_rate": 4.25433037600338e-05,
|
432 |
+
"loss": 2.5018,
|
433 |
+
"step": 6800
|
434 |
+
},
|
435 |
+
{
|
436 |
+
"epoch": 2.92,
|
437 |
+
"learning_rate": 4.1698352344740176e-05,
|
438 |
+
"loss": 2.485,
|
439 |
+
"step": 6900
|
440 |
+
},
|
441 |
+
{
|
442 |
+
"epoch": 2.96,
|
443 |
+
"learning_rate": 4.085340092944656e-05,
|
444 |
+
"loss": 2.5153,
|
445 |
+
"step": 7000
|
446 |
+
},
|
447 |
+
{
|
448 |
+
"epoch": 3.0,
|
449 |
+
"learning_rate": 4.000844951415294e-05,
|
450 |
+
"loss": 2.4789,
|
451 |
+
"step": 7100
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 3.0,
|
455 |
+
"eval_loss": 2.548189401626587,
|
456 |
+
"eval_runtime": 57.5551,
|
457 |
+
"eval_samples_per_second": 185.77,
|
458 |
+
"eval_steps_per_second": 2.919,
|
459 |
+
"step": 7101
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 3.04,
|
463 |
+
"learning_rate": 3.916349809885932e-05,
|
464 |
+
"loss": 2.4652,
|
465 |
+
"step": 7200
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 3.08,
|
469 |
+
"learning_rate": 3.83185466835657e-05,
|
470 |
+
"loss": 2.4474,
|
471 |
+
"step": 7300
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 3.13,
|
475 |
+
"learning_rate": 3.747359526827208e-05,
|
476 |
+
"loss": 2.4374,
|
477 |
+
"step": 7400
|
478 |
+
},
|
479 |
+
{
|
480 |
+
"epoch": 3.17,
|
481 |
+
"learning_rate": 3.6628643852978456e-05,
|
482 |
+
"loss": 2.4465,
|
483 |
+
"step": 7500
|
484 |
+
},
|
485 |
+
{
|
486 |
+
"epoch": 3.21,
|
487 |
+
"learning_rate": 3.5783692437684833e-05,
|
488 |
+
"loss": 2.4391,
|
489 |
+
"step": 7600
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 3.25,
|
493 |
+
"learning_rate": 3.493874102239121e-05,
|
494 |
+
"loss": 2.4286,
|
495 |
+
"step": 7700
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 3.3,
|
499 |
+
"learning_rate": 3.4093789607097596e-05,
|
500 |
+
"loss": 2.4039,
|
501 |
+
"step": 7800
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 3.34,
|
505 |
+
"learning_rate": 3.324883819180397e-05,
|
506 |
+
"loss": 2.4108,
|
507 |
+
"step": 7900
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 3.38,
|
511 |
+
"learning_rate": 3.240388677651035e-05,
|
512 |
+
"loss": 2.4291,
|
513 |
+
"step": 8000
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 3.42,
|
517 |
+
"learning_rate": 3.155893536121673e-05,
|
518 |
+
"loss": 2.4596,
|
519 |
+
"step": 8100
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 3.46,
|
523 |
+
"learning_rate": 3.0713983945923107e-05,
|
524 |
+
"loss": 2.482,
|
525 |
+
"step": 8200
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 3.51,
|
529 |
+
"learning_rate": 2.986903253062949e-05,
|
530 |
+
"loss": 2.444,
|
531 |
+
"step": 8300
|
532 |
+
},
|
533 |
+
{
|
534 |
+
"epoch": 3.55,
|
535 |
+
"learning_rate": 2.9024081115335872e-05,
|
536 |
+
"loss": 2.4132,
|
537 |
+
"step": 8400
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 3.59,
|
541 |
+
"learning_rate": 2.817912970004225e-05,
|
542 |
+
"loss": 2.4506,
|
543 |
+
"step": 8500
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 3.63,
|
547 |
+
"learning_rate": 2.733417828474863e-05,
|
548 |
+
"loss": 2.4385,
|
549 |
+
"step": 8600
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 3.68,
|
553 |
+
"learning_rate": 2.648922686945501e-05,
|
554 |
+
"loss": 2.4386,
|
555 |
+
"step": 8700
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 3.72,
|
559 |
+
"learning_rate": 2.5644275454161386e-05,
|
560 |
+
"loss": 2.434,
|
561 |
+
"step": 8800
|
562 |
+
},
|
563 |
+
{
|
564 |
+
"epoch": 3.76,
|
565 |
+
"learning_rate": 2.4799324038867768e-05,
|
566 |
+
"loss": 2.4477,
|
567 |
+
"step": 8900
|
568 |
+
},
|
569 |
+
{
|
570 |
+
"epoch": 3.8,
|
571 |
+
"learning_rate": 2.3954372623574145e-05,
|
572 |
+
"loss": 2.431,
|
573 |
+
"step": 9000
|
574 |
+
},
|
575 |
+
{
|
576 |
+
"epoch": 3.84,
|
577 |
+
"learning_rate": 2.3109421208280523e-05,
|
578 |
+
"loss": 2.4373,
|
579 |
+
"step": 9100
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 3.89,
|
583 |
+
"learning_rate": 2.2264469792986904e-05,
|
584 |
+
"loss": 2.4193,
|
585 |
+
"step": 9200
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 3.93,
|
589 |
+
"learning_rate": 2.1419518377693285e-05,
|
590 |
+
"loss": 2.4509,
|
591 |
+
"step": 9300
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 3.97,
|
595 |
+
"learning_rate": 2.0574566962399663e-05,
|
596 |
+
"loss": 2.4232,
|
597 |
+
"step": 9400
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 4.0,
|
601 |
+
"eval_loss": 2.529417037963867,
|
602 |
+
"eval_runtime": 57.5729,
|
603 |
+
"eval_samples_per_second": 185.712,
|
604 |
+
"eval_steps_per_second": 2.918,
|
605 |
+
"step": 9468
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 4.01,
|
609 |
+
"learning_rate": 1.9729615547106044e-05,
|
610 |
+
"loss": 2.4341,
|
611 |
+
"step": 9500
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 4.06,
|
615 |
+
"learning_rate": 1.8884664131812422e-05,
|
616 |
+
"loss": 2.3655,
|
617 |
+
"step": 9600
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 4.1,
|
621 |
+
"learning_rate": 1.80397127165188e-05,
|
622 |
+
"loss": 2.4103,
|
623 |
+
"step": 9700
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 4.14,
|
627 |
+
"learning_rate": 1.719476130122518e-05,
|
628 |
+
"loss": 2.4104,
|
629 |
+
"step": 9800
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 4.18,
|
633 |
+
"learning_rate": 1.634980988593156e-05,
|
634 |
+
"loss": 2.4054,
|
635 |
+
"step": 9900
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 4.22,
|
639 |
+
"learning_rate": 1.550485847063794e-05,
|
640 |
+
"loss": 2.4206,
|
641 |
+
"step": 10000
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 4.27,
|
645 |
+
"learning_rate": 1.4659907055344319e-05,
|
646 |
+
"loss": 2.3937,
|
647 |
+
"step": 10100
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 4.31,
|
651 |
+
"learning_rate": 1.3814955640050698e-05,
|
652 |
+
"loss": 2.3804,
|
653 |
+
"step": 10200
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 4.35,
|
657 |
+
"learning_rate": 1.2970004224757076e-05,
|
658 |
+
"loss": 2.3741,
|
659 |
+
"step": 10300
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 4.39,
|
663 |
+
"learning_rate": 1.2125052809463457e-05,
|
664 |
+
"loss": 2.4139,
|
665 |
+
"step": 10400
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 4.44,
|
669 |
+
"learning_rate": 1.1280101394169835e-05,
|
670 |
+
"loss": 2.3955,
|
671 |
+
"step": 10500
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 4.48,
|
675 |
+
"learning_rate": 1.0435149978876214e-05,
|
676 |
+
"loss": 2.3941,
|
677 |
+
"step": 10600
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 4.52,
|
681 |
+
"learning_rate": 9.590198563582595e-06,
|
682 |
+
"loss": 2.4232,
|
683 |
+
"step": 10700
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 4.56,
|
687 |
+
"learning_rate": 8.745247148288973e-06,
|
688 |
+
"loss": 2.396,
|
689 |
+
"step": 10800
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 4.6,
|
693 |
+
"learning_rate": 7.900295732995352e-06,
|
694 |
+
"loss": 2.4021,
|
695 |
+
"step": 10900
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 4.65,
|
699 |
+
"learning_rate": 7.055344317701733e-06,
|
700 |
+
"loss": 2.3959,
|
701 |
+
"step": 11000
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 4.69,
|
705 |
+
"learning_rate": 6.210392902408112e-06,
|
706 |
+
"loss": 2.391,
|
707 |
+
"step": 11100
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 4.73,
|
711 |
+
"learning_rate": 5.3654414871144915e-06,
|
712 |
+
"loss": 2.3885,
|
713 |
+
"step": 11200
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 4.77,
|
717 |
+
"learning_rate": 4.52049007182087e-06,
|
718 |
+
"loss": 2.3712,
|
719 |
+
"step": 11300
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 4.82,
|
723 |
+
"learning_rate": 3.67553865652725e-06,
|
724 |
+
"loss": 2.4148,
|
725 |
+
"step": 11400
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 4.86,
|
729 |
+
"learning_rate": 2.830587241233629e-06,
|
730 |
+
"loss": 2.3986,
|
731 |
+
"step": 11500
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 4.9,
|
735 |
+
"learning_rate": 1.9856358259400087e-06,
|
736 |
+
"loss": 2.3777,
|
737 |
+
"step": 11600
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 4.94,
|
741 |
+
"learning_rate": 1.140684410646388e-06,
|
742 |
+
"loss": 2.3566,
|
743 |
+
"step": 11700
|
744 |
+
},
|
745 |
+
{
|
746 |
+
"epoch": 4.99,
|
747 |
+
"learning_rate": 2.957329953527672e-07,
|
748 |
+
"loss": 2.3971,
|
749 |
+
"step": 11800
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 5.0,
|
753 |
+
"eval_loss": 2.5229861736297607,
|
754 |
+
"eval_runtime": 57.5558,
|
755 |
+
"eval_samples_per_second": 185.768,
|
756 |
+
"eval_steps_per_second": 2.919,
|
757 |
+
"step": 11835
|
758 |
+
},
|
759 |
+
{
|
760 |
+
"epoch": 5.0,
|
761 |
+
"step": 11835,
|
762 |
+
"total_flos": 7.09435871157289e+17,
|
763 |
+
"train_loss": 2.6363602522611114,
|
764 |
+
"train_runtime": 3204.9921,
|
765 |
+
"train_samples_per_second": 118.131,
|
766 |
+
"train_steps_per_second": 3.693
|
767 |
+
}
|
768 |
+
],
|
769 |
+
"logging_steps": 100,
|
770 |
+
"max_steps": 11835,
|
771 |
+
"num_train_epochs": 5,
|
772 |
+
"save_steps": 500,
|
773 |
+
"total_flos": 7.09435871157289e+17,
|
774 |
+
"trial_name": null,
|
775 |
+
"trial_params": null
|
776 |
+
}
|