codestylist
commited on
Commit
•
9b5f408
1
Parent(s):
036a5cf
Upload trainer_state.json
Browse files- trainer_state.json +537 -101
trainer_state.json
CHANGED
@@ -1,184 +1,620 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
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.
|
12 |
-
"learning_rate": 0.
|
13 |
-
"loss":
|
14 |
"step": 500
|
15 |
},
|
16 |
{
|
17 |
-
"epoch": 0.
|
18 |
-
"learning_rate": 0.
|
19 |
-
"loss":
|
20 |
"step": 1000
|
21 |
},
|
22 |
{
|
23 |
-
"epoch": 0.
|
24 |
-
"learning_rate": 0.
|
25 |
-
"loss": 0.
|
26 |
"step": 1500
|
27 |
},
|
28 |
{
|
29 |
-
"epoch": 0.
|
30 |
-
"learning_rate": 0.
|
31 |
-
"loss": 0.
|
32 |
"step": 2000
|
33 |
},
|
34 |
{
|
35 |
-
"epoch": 0.
|
36 |
-
"learning_rate": 0.
|
37 |
-
"loss": 0.
|
38 |
"step": 2500
|
39 |
},
|
40 |
{
|
41 |
-
"epoch": 0.
|
42 |
-
"learning_rate": 0.
|
43 |
-
"loss": 0.
|
44 |
"step": 3000
|
45 |
},
|
46 |
{
|
47 |
-
"epoch":
|
48 |
-
"
|
49 |
-
"
|
50 |
-
"eval_samples_per_second": 178.45,
|
51 |
-
"eval_steps_per_second": 11.163,
|
52 |
-
"step": 3070
|
53 |
-
},
|
54 |
-
{
|
55 |
-
"epoch": 1.14,
|
56 |
-
"learning_rate": 0.0014299674267100977,
|
57 |
-
"loss": 0.0346,
|
58 |
"step": 3500
|
59 |
},
|
60 |
{
|
61 |
-
"epoch":
|
62 |
-
"learning_rate": 0.
|
63 |
-
"loss": 0.
|
64 |
"step": 4000
|
65 |
},
|
66 |
{
|
67 |
-
"epoch":
|
68 |
-
"learning_rate": 0.
|
69 |
-
"loss": 0.
|
70 |
"step": 4500
|
71 |
},
|
72 |
{
|
73 |
-
"epoch":
|
74 |
-
"learning_rate": 0.
|
75 |
-
"loss": 0.
|
76 |
"step": 5000
|
77 |
},
|
78 |
{
|
79 |
-
"epoch":
|
80 |
-
"learning_rate": 0.
|
81 |
-
"loss": 0.
|
82 |
"step": 5500
|
83 |
},
|
84 |
{
|
85 |
-
"epoch":
|
86 |
-
"learning_rate": 0.
|
87 |
-
"loss": 0.
|
88 |
"step": 6000
|
89 |
},
|
90 |
{
|
91 |
-
"epoch":
|
92 |
-
"
|
93 |
-
"
|
94 |
-
"eval_samples_per_second": 178.609,
|
95 |
-
"eval_steps_per_second": 11.173,
|
96 |
-
"step": 6140
|
97 |
-
},
|
98 |
-
{
|
99 |
-
"epoch": 2.12,
|
100 |
-
"learning_rate": 0.0009413680781758957,
|
101 |
-
"loss": 0.0261,
|
102 |
"step": 6500
|
103 |
},
|
104 |
{
|
105 |
-
"epoch":
|
106 |
-
"learning_rate": 0.
|
107 |
-
"loss": 0.
|
108 |
"step": 7000
|
109 |
},
|
110 |
{
|
111 |
-
"epoch":
|
112 |
-
"learning_rate": 0.
|
113 |
-
"loss": 0.
|
114 |
"step": 7500
|
115 |
},
|
116 |
{
|
117 |
-
"epoch":
|
118 |
-
"learning_rate": 0.
|
119 |
-
"loss": 0.
|
120 |
"step": 8000
|
121 |
},
|
122 |
{
|
123 |
-
"epoch":
|
124 |
-
"learning_rate": 0.
|
125 |
-
"loss": 0.
|
126 |
"step": 8500
|
127 |
},
|
128 |
{
|
129 |
-
"epoch":
|
130 |
-
"learning_rate": 0.
|
131 |
-
"loss": 0.
|
132 |
"step": 9000
|
133 |
},
|
134 |
{
|
135 |
-
"epoch":
|
136 |
-
"
|
137 |
-
"
|
138 |
-
"eval_samples_per_second": 178.472,
|
139 |
-
"eval_steps_per_second": 11.165,
|
140 |
-
"step": 9210
|
141 |
-
},
|
142 |
-
{
|
143 |
-
"epoch": 3.09,
|
144 |
-
"learning_rate": 0.0004527687296416938,
|
145 |
-
"loss": 0.0198,
|
146 |
"step": 9500
|
147 |
},
|
148 |
{
|
149 |
-
"epoch":
|
150 |
-
"learning_rate": 0.
|
151 |
-
"loss": 0.
|
152 |
"step": 10000
|
153 |
},
|
154 |
{
|
155 |
-
"epoch":
|
156 |
-
"learning_rate": 0.
|
157 |
-
"loss": 0.
|
158 |
"step": 10500
|
159 |
},
|
160 |
{
|
161 |
-
"epoch":
|
162 |
-
"learning_rate": 0.
|
163 |
-
"loss": 0.
|
164 |
"step": 11000
|
165 |
},
|
166 |
{
|
167 |
-
"epoch":
|
168 |
-
"learning_rate": 0.
|
169 |
-
"loss": 0.
|
170 |
"step": 11500
|
171 |
},
|
172 |
{
|
173 |
-
"epoch":
|
174 |
-
"learning_rate":
|
175 |
-
"loss": 0.
|
176 |
"step": 12000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
177 |
}
|
178 |
],
|
179 |
-
"max_steps":
|
180 |
-
"num_train_epochs":
|
181 |
-
"total_flos":
|
182 |
"trial_name": null,
|
183 |
"trial_params": null
|
184 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.9982255399865387,
|
5 |
+
"global_step": 49000,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 0.001979603907891248,
|
13 |
+
"loss": 0.0528,
|
14 |
"step": 500
|
15 |
},
|
16 |
{
|
17 |
+
"epoch": 0.06,
|
18 |
+
"learning_rate": 0.0019592078157824964,
|
19 |
+
"loss": 0.0543,
|
20 |
"step": 1000
|
21 |
},
|
22 |
{
|
23 |
+
"epoch": 0.09,
|
24 |
+
"learning_rate": 0.0019388117236737441,
|
25 |
+
"loss": 0.0531,
|
26 |
"step": 1500
|
27 |
},
|
28 |
{
|
29 |
+
"epoch": 0.12,
|
30 |
+
"learning_rate": 0.001918415631564992,
|
31 |
+
"loss": 0.0534,
|
32 |
"step": 2000
|
33 |
},
|
34 |
{
|
35 |
+
"epoch": 0.15,
|
36 |
+
"learning_rate": 0.0018980195394562402,
|
37 |
+
"loss": 0.0528,
|
38 |
"step": 2500
|
39 |
},
|
40 |
{
|
41 |
+
"epoch": 0.18,
|
42 |
+
"learning_rate": 0.0018776234473474882,
|
43 |
+
"loss": 0.052,
|
44 |
"step": 3000
|
45 |
},
|
46 |
{
|
47 |
+
"epoch": 0.21,
|
48 |
+
"learning_rate": 0.0018572273552387361,
|
49 |
+
"loss": 0.0518,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
50 |
"step": 3500
|
51 |
},
|
52 |
{
|
53 |
+
"epoch": 0.24,
|
54 |
+
"learning_rate": 0.0018368312631299843,
|
55 |
+
"loss": 0.0509,
|
56 |
"step": 4000
|
57 |
},
|
58 |
{
|
59 |
+
"epoch": 0.28,
|
60 |
+
"learning_rate": 0.0018164351710212323,
|
61 |
+
"loss": 0.0507,
|
62 |
"step": 4500
|
63 |
},
|
64 |
{
|
65 |
+
"epoch": 0.31,
|
66 |
+
"learning_rate": 0.0017960390789124802,
|
67 |
+
"loss": 0.049,
|
68 |
"step": 5000
|
69 |
},
|
70 |
{
|
71 |
+
"epoch": 0.34,
|
72 |
+
"learning_rate": 0.0017756429868037286,
|
73 |
+
"loss": 0.0487,
|
74 |
"step": 5500
|
75 |
},
|
76 |
{
|
77 |
+
"epoch": 0.37,
|
78 |
+
"learning_rate": 0.0017552468946949765,
|
79 |
+
"loss": 0.0484,
|
80 |
"step": 6000
|
81 |
},
|
82 |
{
|
83 |
+
"epoch": 0.4,
|
84 |
+
"learning_rate": 0.0017348508025862245,
|
85 |
+
"loss": 0.0498,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
86 |
"step": 6500
|
87 |
},
|
88 |
{
|
89 |
+
"epoch": 0.43,
|
90 |
+
"learning_rate": 0.0017144547104774727,
|
91 |
+
"loss": 0.0507,
|
92 |
"step": 7000
|
93 |
},
|
94 |
{
|
95 |
+
"epoch": 0.46,
|
96 |
+
"learning_rate": 0.0016940586183687206,
|
97 |
+
"loss": 0.0467,
|
98 |
"step": 7500
|
99 |
},
|
100 |
{
|
101 |
+
"epoch": 0.49,
|
102 |
+
"learning_rate": 0.0016736625262599688,
|
103 |
+
"loss": 0.0469,
|
104 |
"step": 8000
|
105 |
},
|
106 |
{
|
107 |
+
"epoch": 0.52,
|
108 |
+
"learning_rate": 0.0016532664341512167,
|
109 |
+
"loss": 0.0473,
|
110 |
"step": 8500
|
111 |
},
|
112 |
{
|
113 |
+
"epoch": 0.55,
|
114 |
+
"learning_rate": 0.0016328703420424647,
|
115 |
+
"loss": 0.047,
|
116 |
"step": 9000
|
117 |
},
|
118 |
{
|
119 |
+
"epoch": 0.58,
|
120 |
+
"learning_rate": 0.0016124742499337129,
|
121 |
+
"loss": 0.0465,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
122 |
"step": 9500
|
123 |
},
|
124 |
{
|
125 |
+
"epoch": 0.61,
|
126 |
+
"learning_rate": 0.0015920781578249608,
|
127 |
+
"loss": 0.0461,
|
128 |
"step": 10000
|
129 |
},
|
130 |
{
|
131 |
+
"epoch": 0.64,
|
132 |
+
"learning_rate": 0.0015716820657162088,
|
133 |
+
"loss": 0.045,
|
134 |
"step": 10500
|
135 |
},
|
136 |
{
|
137 |
+
"epoch": 0.67,
|
138 |
+
"learning_rate": 0.001551285973607457,
|
139 |
+
"loss": 0.0468,
|
140 |
"step": 11000
|
141 |
},
|
142 |
{
|
143 |
+
"epoch": 0.7,
|
144 |
+
"learning_rate": 0.0015308898814987049,
|
145 |
+
"loss": 0.0445,
|
146 |
"step": 11500
|
147 |
},
|
148 |
{
|
149 |
+
"epoch": 0.73,
|
150 |
+
"learning_rate": 0.0015104937893899528,
|
151 |
+
"loss": 0.0443,
|
152 |
"step": 12000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.76,
|
156 |
+
"learning_rate": 0.001490097697281201,
|
157 |
+
"loss": 0.0439,
|
158 |
+
"step": 12500
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.8,
|
162 |
+
"learning_rate": 0.001469701605172449,
|
163 |
+
"loss": 0.0437,
|
164 |
+
"step": 13000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.83,
|
168 |
+
"learning_rate": 0.001449305513063697,
|
169 |
+
"loss": 0.0432,
|
170 |
+
"step": 13500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.86,
|
174 |
+
"learning_rate": 0.001428909420954945,
|
175 |
+
"loss": 0.0434,
|
176 |
+
"step": 14000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.89,
|
180 |
+
"learning_rate": 0.001408513328846193,
|
181 |
+
"loss": 0.044,
|
182 |
+
"step": 14500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.92,
|
186 |
+
"learning_rate": 0.001388117236737441,
|
187 |
+
"loss": 0.0433,
|
188 |
+
"step": 15000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.95,
|
192 |
+
"learning_rate": 0.0013677211446286891,
|
193 |
+
"loss": 0.0427,
|
194 |
+
"step": 15500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.98,
|
198 |
+
"learning_rate": 0.001347325052519937,
|
199 |
+
"loss": 0.043,
|
200 |
+
"step": 16000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 1.0,
|
204 |
+
"eval_loss": 0.03870956972241402,
|
205 |
+
"eval_runtime": 667.1114,
|
206 |
+
"eval_samples_per_second": 146.987,
|
207 |
+
"eval_steps_per_second": 6.125,
|
208 |
+
"step": 16343
|
209 |
+
},
|
210 |
+
{
|
211 |
+
"epoch": 1.01,
|
212 |
+
"learning_rate": 0.0013269289604111853,
|
213 |
+
"loss": 0.0404,
|
214 |
+
"step": 16500
|
215 |
+
},
|
216 |
+
{
|
217 |
+
"epoch": 1.04,
|
218 |
+
"learning_rate": 0.0013065328683024334,
|
219 |
+
"loss": 0.037,
|
220 |
+
"step": 17000
|
221 |
+
},
|
222 |
+
{
|
223 |
+
"epoch": 1.07,
|
224 |
+
"learning_rate": 0.0012861367761936814,
|
225 |
+
"loss": 0.0371,
|
226 |
+
"step": 17500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 1.1,
|
230 |
+
"learning_rate": 0.0012657406840849293,
|
231 |
+
"loss": 0.0372,
|
232 |
+
"step": 18000
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 1.13,
|
236 |
+
"learning_rate": 0.0012453445919761775,
|
237 |
+
"loss": 0.0386,
|
238 |
+
"step": 18500
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 1.16,
|
242 |
+
"learning_rate": 0.0012249484998674255,
|
243 |
+
"loss": 0.0377,
|
244 |
+
"step": 19000
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 1.19,
|
248 |
+
"learning_rate": 0.0012045524077586734,
|
249 |
+
"loss": 0.0385,
|
250 |
+
"step": 19500
|
251 |
+
},
|
252 |
+
{
|
253 |
+
"epoch": 1.22,
|
254 |
+
"learning_rate": 0.0011841563156499216,
|
255 |
+
"loss": 0.0377,
|
256 |
+
"step": 20000
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 1.25,
|
260 |
+
"learning_rate": 0.0011637602235411695,
|
261 |
+
"loss": 0.0378,
|
262 |
+
"step": 20500
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 1.28,
|
266 |
+
"learning_rate": 0.0011433641314324175,
|
267 |
+
"loss": 0.0364,
|
268 |
+
"step": 21000
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 1.32,
|
272 |
+
"learning_rate": 0.0011229680393236656,
|
273 |
+
"loss": 0.0372,
|
274 |
+
"step": 21500
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 1.35,
|
278 |
+
"learning_rate": 0.0011025719472149136,
|
279 |
+
"loss": 0.0372,
|
280 |
+
"step": 22000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 1.38,
|
284 |
+
"learning_rate": 0.0010821758551061618,
|
285 |
+
"loss": 0.0387,
|
286 |
+
"step": 22500
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 1.41,
|
290 |
+
"learning_rate": 0.0010617797629974097,
|
291 |
+
"loss": 0.0387,
|
292 |
+
"step": 23000
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 1.44,
|
296 |
+
"learning_rate": 0.0010413836708886577,
|
297 |
+
"loss": 0.0387,
|
298 |
+
"step": 23500
|
299 |
+
},
|
300 |
+
{
|
301 |
+
"epoch": 1.47,
|
302 |
+
"learning_rate": 0.0010209875787799058,
|
303 |
+
"loss": 0.0363,
|
304 |
+
"step": 24000
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.5,
|
308 |
+
"learning_rate": 0.0010005914866711538,
|
309 |
+
"loss": 0.0385,
|
310 |
+
"step": 24500
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 1.53,
|
314 |
+
"learning_rate": 0.0009801953945624017,
|
315 |
+
"loss": 0.0387,
|
316 |
+
"step": 25000
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.56,
|
320 |
+
"learning_rate": 0.0009597993024536499,
|
321 |
+
"loss": 0.0381,
|
322 |
+
"step": 25500
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 1.59,
|
326 |
+
"learning_rate": 0.000939403210344898,
|
327 |
+
"loss": 0.039,
|
328 |
+
"step": 26000
|
329 |
+
},
|
330 |
+
{
|
331 |
+
"epoch": 1.62,
|
332 |
+
"learning_rate": 0.000919007118236146,
|
333 |
+
"loss": 0.0369,
|
334 |
+
"step": 26500
|
335 |
+
},
|
336 |
+
{
|
337 |
+
"epoch": 1.65,
|
338 |
+
"learning_rate": 0.000898611026127394,
|
339 |
+
"loss": 0.0372,
|
340 |
+
"step": 27000
|
341 |
+
},
|
342 |
+
{
|
343 |
+
"epoch": 1.68,
|
344 |
+
"learning_rate": 0.000878214934018642,
|
345 |
+
"loss": 0.0379,
|
346 |
+
"step": 27500
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.71,
|
350 |
+
"learning_rate": 0.0008578188419098901,
|
351 |
+
"loss": 0.0367,
|
352 |
+
"step": 28000
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 1.74,
|
356 |
+
"learning_rate": 0.0008374227498011381,
|
357 |
+
"loss": 0.0361,
|
358 |
+
"step": 28500
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 1.77,
|
362 |
+
"learning_rate": 0.0008170266576923861,
|
363 |
+
"loss": 0.0364,
|
364 |
+
"step": 29000
|
365 |
+
},
|
366 |
+
{
|
367 |
+
"epoch": 1.81,
|
368 |
+
"learning_rate": 0.0007966305655836342,
|
369 |
+
"loss": 0.0361,
|
370 |
+
"step": 29500
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 1.84,
|
374 |
+
"learning_rate": 0.0007762344734748822,
|
375 |
+
"loss": 0.0359,
|
376 |
+
"step": 30000
|
377 |
+
},
|
378 |
+
{
|
379 |
+
"epoch": 1.87,
|
380 |
+
"learning_rate": 0.0007558383813661303,
|
381 |
+
"loss": 0.0357,
|
382 |
+
"step": 30500
|
383 |
+
},
|
384 |
+
{
|
385 |
+
"epoch": 1.9,
|
386 |
+
"learning_rate": 0.0007354422892573784,
|
387 |
+
"loss": 0.0358,
|
388 |
+
"step": 31000
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 1.93,
|
392 |
+
"learning_rate": 0.0007150461971486263,
|
393 |
+
"loss": 0.0357,
|
394 |
+
"step": 31500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 1.96,
|
398 |
+
"learning_rate": 0.0006946501050398744,
|
399 |
+
"loss": 0.0352,
|
400 |
+
"step": 32000
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 1.99,
|
404 |
+
"learning_rate": 0.0006742540129311224,
|
405 |
+
"loss": 0.0351,
|
406 |
+
"step": 32500
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 2.0,
|
410 |
+
"eval_loss": 0.03284740820527077,
|
411 |
+
"eval_runtime": 666.0258,
|
412 |
+
"eval_samples_per_second": 147.227,
|
413 |
+
"eval_steps_per_second": 6.135,
|
414 |
+
"step": 32686
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 2.02,
|
418 |
+
"learning_rate": 0.0006538579208223705,
|
419 |
+
"loss": 0.0325,
|
420 |
+
"step": 33000
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 2.05,
|
424 |
+
"learning_rate": 0.0006334618287136184,
|
425 |
+
"loss": 0.0314,
|
426 |
+
"step": 33500
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 2.08,
|
430 |
+
"learning_rate": 0.0006130657366048665,
|
431 |
+
"loss": 0.0308,
|
432 |
+
"step": 34000
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 2.11,
|
436 |
+
"learning_rate": 0.0005926696444961146,
|
437 |
+
"loss": 0.0315,
|
438 |
+
"step": 34500
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 2.14,
|
442 |
+
"learning_rate": 0.0005722735523873625,
|
443 |
+
"loss": 0.0308,
|
444 |
+
"step": 35000
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 2.17,
|
448 |
+
"learning_rate": 0.0005518774602786107,
|
449 |
+
"loss": 0.0306,
|
450 |
+
"step": 35500
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 2.2,
|
454 |
+
"learning_rate": 0.0005314813681698587,
|
455 |
+
"loss": 0.0309,
|
456 |
+
"step": 36000
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 2.23,
|
460 |
+
"learning_rate": 0.0005110852760611067,
|
461 |
+
"loss": 0.0304,
|
462 |
+
"step": 36500
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 2.26,
|
466 |
+
"learning_rate": 0.0004906891839523547,
|
467 |
+
"loss": 0.0306,
|
468 |
+
"step": 37000
|
469 |
+
},
|
470 |
+
{
|
471 |
+
"epoch": 2.29,
|
472 |
+
"learning_rate": 0.00047029309184360276,
|
473 |
+
"loss": 0.0307,
|
474 |
+
"step": 37500
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 2.33,
|
478 |
+
"learning_rate": 0.0004498969997348508,
|
479 |
+
"loss": 0.0296,
|
480 |
+
"step": 38000
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 2.36,
|
484 |
+
"learning_rate": 0.0004295009076260988,
|
485 |
+
"loss": 0.0302,
|
486 |
+
"step": 38500
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 2.39,
|
490 |
+
"learning_rate": 0.0004091048155173469,
|
491 |
+
"loss": 0.0299,
|
492 |
+
"step": 39000
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 2.42,
|
496 |
+
"learning_rate": 0.00038870872340859494,
|
497 |
+
"loss": 0.0297,
|
498 |
+
"step": 39500
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 2.45,
|
502 |
+
"learning_rate": 0.00036831263129984295,
|
503 |
+
"loss": 0.0292,
|
504 |
+
"step": 40000
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 2.48,
|
508 |
+
"learning_rate": 0.000347916539191091,
|
509 |
+
"loss": 0.0302,
|
510 |
+
"step": 40500
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 2.51,
|
514 |
+
"learning_rate": 0.000327520447082339,
|
515 |
+
"loss": 0.0295,
|
516 |
+
"step": 41000
|
517 |
+
},
|
518 |
+
{
|
519 |
+
"epoch": 2.54,
|
520 |
+
"learning_rate": 0.0003071243549735871,
|
521 |
+
"loss": 0.0295,
|
522 |
+
"step": 41500
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 2.57,
|
526 |
+
"learning_rate": 0.00028672826286483513,
|
527 |
+
"loss": 0.0289,
|
528 |
+
"step": 42000
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 2.6,
|
532 |
+
"learning_rate": 0.00026633217075608314,
|
533 |
+
"loss": 0.0297,
|
534 |
+
"step": 42500
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 2.63,
|
538 |
+
"learning_rate": 0.00024593607864733115,
|
539 |
+
"loss": 0.0291,
|
540 |
+
"step": 43000
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 2.66,
|
544 |
+
"learning_rate": 0.0002255399865385792,
|
545 |
+
"loss": 0.0288,
|
546 |
+
"step": 43500
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 2.69,
|
550 |
+
"learning_rate": 0.00020514389442982727,
|
551 |
+
"loss": 0.0278,
|
552 |
+
"step": 44000
|
553 |
+
},
|
554 |
+
{
|
555 |
+
"epoch": 2.72,
|
556 |
+
"learning_rate": 0.0001847478023210753,
|
557 |
+
"loss": 0.0284,
|
558 |
+
"step": 44500
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 2.75,
|
562 |
+
"learning_rate": 0.0001643517102123233,
|
563 |
+
"loss": 0.0285,
|
564 |
+
"step": 45000
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 2.78,
|
568 |
+
"learning_rate": 0.00014395561810357136,
|
569 |
+
"loss": 0.0278,
|
570 |
+
"step": 45500
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 2.81,
|
574 |
+
"learning_rate": 0.0001235595259948194,
|
575 |
+
"loss": 0.0286,
|
576 |
+
"step": 46000
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 2.85,
|
580 |
+
"learning_rate": 0.00010316343388606743,
|
581 |
+
"loss": 0.028,
|
582 |
+
"step": 46500
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 2.88,
|
586 |
+
"learning_rate": 8.276734177731548e-05,
|
587 |
+
"loss": 0.0279,
|
588 |
+
"step": 47000
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 2.91,
|
592 |
+
"learning_rate": 6.23712496685635e-05,
|
593 |
+
"loss": 0.0286,
|
594 |
+
"step": 47500
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 2.94,
|
598 |
+
"learning_rate": 4.197515755981154e-05,
|
599 |
+
"loss": 0.0278,
|
600 |
+
"step": 48000
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 2.97,
|
604 |
+
"learning_rate": 2.157906545105958e-05,
|
605 |
+
"loss": 0.0279,
|
606 |
+
"step": 48500
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 3.0,
|
610 |
+
"learning_rate": 1.1829733423076138e-06,
|
611 |
+
"loss": 0.0277,
|
612 |
+
"step": 49000
|
613 |
}
|
614 |
],
|
615 |
+
"max_steps": 49029,
|
616 |
+
"num_train_epochs": 3,
|
617 |
+
"total_flos": 1.591600637458514e+17,
|
618 |
"trial_name": null,
|
619 |
"trial_params": null
|
620 |
}
|