Upload trainer_state.json
Browse files- trainer_state.json +259 -0
trainer_state.json
ADDED
@@ -0,0 +1,259 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 35.0,
|
5 |
+
"global_step": 19705,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.89,
|
12 |
+
"learning_rate": 4.8731286475513835e-05,
|
13 |
+
"loss": 3.7437,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 1.78,
|
18 |
+
"learning_rate": 4.746257295102766e-05,
|
19 |
+
"loss": 3.203,
|
20 |
+
"step": 1000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 2.66,
|
24 |
+
"learning_rate": 4.619385942654149e-05,
|
25 |
+
"loss": 2.9759,
|
26 |
+
"step": 1500
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 3.55,
|
30 |
+
"learning_rate": 4.492514590205532e-05,
|
31 |
+
"loss": 2.7917,
|
32 |
+
"step": 2000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 4.44,
|
36 |
+
"learning_rate": 4.365643237756914e-05,
|
37 |
+
"loss": 2.6427,
|
38 |
+
"step": 2500
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 5.33,
|
42 |
+
"learning_rate": 4.2387718853082975e-05,
|
43 |
+
"loss": 2.5541,
|
44 |
+
"step": 3000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 6.22,
|
48 |
+
"learning_rate": 4.111900532859681e-05,
|
49 |
+
"loss": 2.4485,
|
50 |
+
"step": 3500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 7.1,
|
54 |
+
"learning_rate": 3.985029180411063e-05,
|
55 |
+
"loss": 2.3436,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 7.99,
|
60 |
+
"learning_rate": 3.8581578279624465e-05,
|
61 |
+
"loss": 2.2573,
|
62 |
+
"step": 4500
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 8.88,
|
66 |
+
"learning_rate": 3.73128647551383e-05,
|
67 |
+
"loss": 2.1701,
|
68 |
+
"step": 5000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 9.77,
|
72 |
+
"learning_rate": 3.6044151230652115e-05,
|
73 |
+
"loss": 2.0915,
|
74 |
+
"step": 5500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 10.66,
|
78 |
+
"learning_rate": 3.477543770616595e-05,
|
79 |
+
"loss": 2.0424,
|
80 |
+
"step": 6000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 11.55,
|
84 |
+
"learning_rate": 3.350672418167978e-05,
|
85 |
+
"loss": 1.9703,
|
86 |
+
"step": 6500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 12.43,
|
90 |
+
"learning_rate": 3.2238010657193605e-05,
|
91 |
+
"loss": 1.9065,
|
92 |
+
"step": 7000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 13.32,
|
96 |
+
"learning_rate": 3.096929713270744e-05,
|
97 |
+
"loss": 1.877,
|
98 |
+
"step": 7500
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 14.21,
|
102 |
+
"learning_rate": 2.9700583608221266e-05,
|
103 |
+
"loss": 1.8305,
|
104 |
+
"step": 8000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 15.1,
|
108 |
+
"learning_rate": 2.8431870083735095e-05,
|
109 |
+
"loss": 1.7644,
|
110 |
+
"step": 8500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 15.99,
|
114 |
+
"learning_rate": 2.7163156559248927e-05,
|
115 |
+
"loss": 1.7179,
|
116 |
+
"step": 9000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 16.87,
|
120 |
+
"learning_rate": 2.589444303476275e-05,
|
121 |
+
"loss": 1.6708,
|
122 |
+
"step": 9500
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 17.76,
|
126 |
+
"learning_rate": 2.462572951027658e-05,
|
127 |
+
"loss": 1.6338,
|
128 |
+
"step": 10000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 18.65,
|
132 |
+
"learning_rate": 2.335701598579041e-05,
|
133 |
+
"loss": 1.5861,
|
134 |
+
"step": 10500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 19.54,
|
138 |
+
"learning_rate": 2.208830246130424e-05,
|
139 |
+
"loss": 1.553,
|
140 |
+
"step": 11000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 20.43,
|
144 |
+
"learning_rate": 2.0819588936818067e-05,
|
145 |
+
"loss": 1.538,
|
146 |
+
"step": 11500
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 21.31,
|
150 |
+
"learning_rate": 1.9550875412331896e-05,
|
151 |
+
"loss": 1.5036,
|
152 |
+
"step": 12000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 22.2,
|
156 |
+
"learning_rate": 1.8282161887845725e-05,
|
157 |
+
"loss": 1.4526,
|
158 |
+
"step": 12500
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 23.09,
|
162 |
+
"learning_rate": 1.7013448363359553e-05,
|
163 |
+
"loss": 1.4547,
|
164 |
+
"step": 13000
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 23.98,
|
168 |
+
"learning_rate": 1.5744734838873386e-05,
|
169 |
+
"loss": 1.4221,
|
170 |
+
"step": 13500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 24.87,
|
174 |
+
"learning_rate": 1.4476021314387211e-05,
|
175 |
+
"loss": 1.3794,
|
176 |
+
"step": 14000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 25.75,
|
180 |
+
"learning_rate": 1.3207307789901041e-05,
|
181 |
+
"loss": 1.3785,
|
182 |
+
"step": 14500
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 26.64,
|
186 |
+
"learning_rate": 1.193859426541487e-05,
|
187 |
+
"loss": 1.345,
|
188 |
+
"step": 15000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 27.53,
|
192 |
+
"learning_rate": 1.0669880740928699e-05,
|
193 |
+
"loss": 1.326,
|
194 |
+
"step": 15500
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 28.42,
|
198 |
+
"learning_rate": 9.401167216442528e-06,
|
199 |
+
"loss": 1.3241,
|
200 |
+
"step": 16000
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 29.31,
|
204 |
+
"learning_rate": 8.132453691956356e-06,
|
205 |
+
"loss": 1.2962,
|
206 |
+
"step": 16500
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 30.2,
|
210 |
+
"learning_rate": 6.863740167470185e-06,
|
211 |
+
"loss": 1.2871,
|
212 |
+
"step": 17000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 31.08,
|
216 |
+
"learning_rate": 5.595026642984015e-06,
|
217 |
+
"loss": 1.2761,
|
218 |
+
"step": 17500
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 31.97,
|
222 |
+
"learning_rate": 4.3263131184978435e-06,
|
223 |
+
"loss": 1.2549,
|
224 |
+
"step": 18000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 32.86,
|
228 |
+
"learning_rate": 3.0575995940116723e-06,
|
229 |
+
"loss": 1.252,
|
230 |
+
"step": 18500
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 33.75,
|
234 |
+
"learning_rate": 1.7888860695255013e-06,
|
235 |
+
"loss": 1.2429,
|
236 |
+
"step": 19000
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 34.64,
|
240 |
+
"learning_rate": 5.201725450393302e-07,
|
241 |
+
"loss": 1.2436,
|
242 |
+
"step": 19500
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 35.0,
|
246 |
+
"step": 19705,
|
247 |
+
"total_flos": 9895714212962304.0,
|
248 |
+
"train_loss": 1.8335362397363058,
|
249 |
+
"train_runtime": 2725.9313,
|
250 |
+
"train_samples_per_second": 216.554,
|
251 |
+
"train_steps_per_second": 7.229
|
252 |
+
}
|
253 |
+
],
|
254 |
+
"max_steps": 19705,
|
255 |
+
"num_train_epochs": 35,
|
256 |
+
"total_flos": 9895714212962304.0,
|
257 |
+
"trial_name": null,
|
258 |
+
"trial_params": null
|
259 |
+
}
|