End of training
Browse files
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.94,
|
3 |
+
"eval_accuracy": 0.9623477297895903,
|
4 |
+
"eval_loss": 0.12883274257183075,
|
5 |
+
"eval_runtime": 141.8294,
|
6 |
+
"eval_samples_per_second": 31.834,
|
7 |
+
"eval_steps_per_second": 1.995,
|
8 |
+
"total_flos": 7.375405476885369e+18,
|
9 |
+
"train_loss": 0.5960829522569425,
|
10 |
+
"train_runtime": 11023.3494,
|
11 |
+
"train_samples_per_second": 3.801,
|
12 |
+
"train_steps_per_second": 0.059
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.94,
|
3 |
+
"eval_accuracy": 0.9623477297895903,
|
4 |
+
"eval_loss": 0.12883274257183075,
|
5 |
+
"eval_runtime": 141.8294,
|
6 |
+
"eval_samples_per_second": 31.834,
|
7 |
+
"eval_steps_per_second": 1.995
|
8 |
+
}
|
runs/Jul23_12-29-02_667be8c9d284/events.out.tfevents.1690126843.667be8c9d284.1693.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:36f03a52d2acb683e959f2ef940826c3b332dc46f1f2a7a7c5f135d2e1e007b3
|
3 |
+
size 779
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.94,
|
3 |
+
"total_flos": 7.375405476885369e+18,
|
4 |
+
"train_loss": 0.5960829522569425,
|
5 |
+
"train_runtime": 11023.3494,
|
6 |
+
"train_samples_per_second": 3.801,
|
7 |
+
"train_steps_per_second": 0.059
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,868 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.9623477297895903,
|
3 |
+
"best_model_checkpoint": "convnextv2-large-1k-224-finetuned-LungCancer-Classification-LC25000-AH-40-30-30-Shuffled/checkpoint-93",
|
4 |
+
"epoch": 6.944,
|
5 |
+
"global_step": 651,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.05,
|
12 |
+
"learning_rate": 7.668711656441718e-06,
|
13 |
+
"loss": 1.0754,
|
14 |
+
"step": 5
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.11,
|
18 |
+
"learning_rate": 1.5337423312883436e-05,
|
19 |
+
"loss": 1.0442,
|
20 |
+
"step": 10
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.16,
|
24 |
+
"learning_rate": 2.300613496932515e-05,
|
25 |
+
"loss": 0.9585,
|
26 |
+
"step": 15
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.21,
|
30 |
+
"learning_rate": 3.067484662576687e-05,
|
31 |
+
"loss": 0.8575,
|
32 |
+
"step": 20
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.27,
|
36 |
+
"learning_rate": 3.834355828220859e-05,
|
37 |
+
"loss": 0.7163,
|
38 |
+
"step": 25
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.32,
|
42 |
+
"learning_rate": 4.60122699386503e-05,
|
43 |
+
"loss": 0.5779,
|
44 |
+
"step": 30
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.37,
|
48 |
+
"learning_rate": 5.368098159509203e-05,
|
49 |
+
"loss": 0.4602,
|
50 |
+
"step": 35
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.43,
|
54 |
+
"learning_rate": 6.134969325153375e-05,
|
55 |
+
"loss": 0.3681,
|
56 |
+
"step": 40
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.48,
|
60 |
+
"learning_rate": 6.901840490797547e-05,
|
61 |
+
"loss": 0.311,
|
62 |
+
"step": 45
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.53,
|
66 |
+
"learning_rate": 7.668711656441718e-05,
|
67 |
+
"loss": 0.2978,
|
68 |
+
"step": 50
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.59,
|
72 |
+
"learning_rate": 8.43558282208589e-05,
|
73 |
+
"loss": 0.2204,
|
74 |
+
"step": 55
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.64,
|
78 |
+
"learning_rate": 9.20245398773006e-05,
|
79 |
+
"loss": 0.2907,
|
80 |
+
"step": 60
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.69,
|
84 |
+
"learning_rate": 9.969325153374233e-05,
|
85 |
+
"loss": 0.2103,
|
86 |
+
"step": 65
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.75,
|
90 |
+
"learning_rate": 0.00010736196319018406,
|
91 |
+
"loss": 0.2288,
|
92 |
+
"step": 70
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.8,
|
96 |
+
"learning_rate": 0.00011503067484662577,
|
97 |
+
"loss": 0.1962,
|
98 |
+
"step": 75
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.85,
|
102 |
+
"learning_rate": 0.0001226993865030675,
|
103 |
+
"loss": 0.2876,
|
104 |
+
"step": 80
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.91,
|
108 |
+
"learning_rate": 0.0001303680981595092,
|
109 |
+
"loss": 0.2127,
|
110 |
+
"step": 85
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.96,
|
114 |
+
"learning_rate": 0.00013803680981595094,
|
115 |
+
"loss": 0.2522,
|
116 |
+
"step": 90
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.99,
|
120 |
+
"eval_accuracy": 0.9623477297895903,
|
121 |
+
"eval_loss": 0.12883274257183075,
|
122 |
+
"eval_runtime": 2650.7364,
|
123 |
+
"eval_samples_per_second": 1.703,
|
124 |
+
"eval_steps_per_second": 0.107,
|
125 |
+
"step": 93
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 1.01,
|
129 |
+
"learning_rate": 0.00014570552147239263,
|
130 |
+
"loss": 0.2536,
|
131 |
+
"step": 95
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 1.07,
|
135 |
+
"learning_rate": 0.00015337423312883436,
|
136 |
+
"loss": 0.2064,
|
137 |
+
"step": 100
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 1.12,
|
141 |
+
"learning_rate": 0.00016104294478527608,
|
142 |
+
"loss": 0.3041,
|
143 |
+
"step": 105
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 1.17,
|
147 |
+
"learning_rate": 0.0001687116564417178,
|
148 |
+
"loss": 0.2007,
|
149 |
+
"step": 110
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.23,
|
153 |
+
"learning_rate": 0.00017638036809815953,
|
154 |
+
"loss": 0.1304,
|
155 |
+
"step": 115
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 1.28,
|
159 |
+
"learning_rate": 0.0001840490797546012,
|
160 |
+
"loss": 0.2096,
|
161 |
+
"step": 120
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 1.33,
|
165 |
+
"learning_rate": 0.00019171779141104295,
|
166 |
+
"loss": 0.1957,
|
167 |
+
"step": 125
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 1.39,
|
171 |
+
"learning_rate": 0.00019938650306748466,
|
172 |
+
"loss": 0.1713,
|
173 |
+
"step": 130
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 1.44,
|
177 |
+
"learning_rate": 0.00020705521472392637,
|
178 |
+
"loss": 0.3216,
|
179 |
+
"step": 135
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 1.49,
|
183 |
+
"learning_rate": 0.0002147239263803681,
|
184 |
+
"loss": 0.3009,
|
185 |
+
"step": 140
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 1.55,
|
189 |
+
"learning_rate": 0.00022239263803680982,
|
190 |
+
"loss": 0.2504,
|
191 |
+
"step": 145
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 1.6,
|
195 |
+
"learning_rate": 0.00023006134969325153,
|
196 |
+
"loss": 0.2467,
|
197 |
+
"step": 150
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 1.65,
|
201 |
+
"learning_rate": 0.00023773006134969325,
|
202 |
+
"loss": 0.5555,
|
203 |
+
"step": 155
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 1.71,
|
207 |
+
"learning_rate": 0.000245398773006135,
|
208 |
+
"loss": 0.2855,
|
209 |
+
"step": 160
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 1.76,
|
213 |
+
"learning_rate": 0.00025306748466257667,
|
214 |
+
"loss": 0.1879,
|
215 |
+
"step": 165
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 1.81,
|
219 |
+
"learning_rate": 0.0002607361963190184,
|
220 |
+
"loss": 0.2202,
|
221 |
+
"step": 170
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 1.87,
|
225 |
+
"learning_rate": 0.00026840490797546015,
|
226 |
+
"loss": 0.2338,
|
227 |
+
"step": 175
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 1.92,
|
231 |
+
"learning_rate": 0.0002760736196319019,
|
232 |
+
"loss": 0.2014,
|
233 |
+
"step": 180
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 1.97,
|
237 |
+
"learning_rate": 0.00028374233128834357,
|
238 |
+
"loss": 0.1579,
|
239 |
+
"step": 185
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 1.99,
|
243 |
+
"eval_accuracy": 0.9572535991140643,
|
244 |
+
"eval_loss": 0.12107633799314499,
|
245 |
+
"eval_runtime": 150.1469,
|
246 |
+
"eval_samples_per_second": 30.071,
|
247 |
+
"eval_steps_per_second": 1.885,
|
248 |
+
"step": 187
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 2.03,
|
252 |
+
"learning_rate": 0.00029141104294478525,
|
253 |
+
"loss": 0.2111,
|
254 |
+
"step": 190
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 2.08,
|
258 |
+
"learning_rate": 0.000299079754601227,
|
259 |
+
"loss": 0.247,
|
260 |
+
"step": 195
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 2.13,
|
264 |
+
"learning_rate": 0.00030674846625766873,
|
265 |
+
"loss": 0.2446,
|
266 |
+
"step": 200
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 2.19,
|
270 |
+
"learning_rate": 0.00031441717791411047,
|
271 |
+
"loss": 0.2666,
|
272 |
+
"step": 205
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 2.24,
|
276 |
+
"learning_rate": 0.00032208588957055215,
|
277 |
+
"loss": 0.1644,
|
278 |
+
"step": 210
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 2.29,
|
282 |
+
"learning_rate": 0.00032975460122699384,
|
283 |
+
"loss": 0.3137,
|
284 |
+
"step": 215
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 2.35,
|
288 |
+
"learning_rate": 0.0003374233128834356,
|
289 |
+
"loss": 0.3366,
|
290 |
+
"step": 220
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 2.4,
|
294 |
+
"learning_rate": 0.0003450920245398773,
|
295 |
+
"loss": 0.2326,
|
296 |
+
"step": 225
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 2.45,
|
300 |
+
"learning_rate": 0.00035276073619631905,
|
301 |
+
"loss": 0.588,
|
302 |
+
"step": 230
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 2.51,
|
306 |
+
"learning_rate": 0.00036042944785276074,
|
307 |
+
"loss": 0.9461,
|
308 |
+
"step": 235
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 2.56,
|
312 |
+
"learning_rate": 0.0003680981595092024,
|
313 |
+
"loss": 1.1656,
|
314 |
+
"step": 240
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 2.61,
|
318 |
+
"learning_rate": 0.00037576687116564416,
|
319 |
+
"loss": 1.3311,
|
320 |
+
"step": 245
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 2.67,
|
324 |
+
"learning_rate": 0.0003834355828220859,
|
325 |
+
"loss": 1.1308,
|
326 |
+
"step": 250
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 2.72,
|
330 |
+
"learning_rate": 0.00039110429447852764,
|
331 |
+
"loss": 1.1135,
|
332 |
+
"step": 255
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 2.77,
|
336 |
+
"learning_rate": 0.0003987730061349693,
|
337 |
+
"loss": 1.1257,
|
338 |
+
"step": 260
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 2.83,
|
342 |
+
"learning_rate": 0.00040644171779141106,
|
343 |
+
"loss": 1.1105,
|
344 |
+
"step": 265
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 2.88,
|
348 |
+
"learning_rate": 0.00041411042944785274,
|
349 |
+
"loss": 1.12,
|
350 |
+
"step": 270
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 2.93,
|
354 |
+
"learning_rate": 0.0004217791411042945,
|
355 |
+
"loss": 1.1094,
|
356 |
+
"step": 275
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 2.99,
|
360 |
+
"learning_rate": 0.0004294478527607362,
|
361 |
+
"loss": 1.1016,
|
362 |
+
"step": 280
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 3.0,
|
366 |
+
"eval_accuracy": 0.3215946843853821,
|
367 |
+
"eval_loss": 1.101806640625,
|
368 |
+
"eval_runtime": 149.0854,
|
369 |
+
"eval_samples_per_second": 30.285,
|
370 |
+
"eval_steps_per_second": 1.898,
|
371 |
+
"step": 281
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 3.04,
|
375 |
+
"learning_rate": 0.0004371165644171779,
|
376 |
+
"loss": 1.0957,
|
377 |
+
"step": 285
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 3.09,
|
381 |
+
"learning_rate": 0.00044478527607361964,
|
382 |
+
"loss": 1.1146,
|
383 |
+
"step": 290
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 3.15,
|
387 |
+
"learning_rate": 0.0004524539877300614,
|
388 |
+
"loss": 1.0941,
|
389 |
+
"step": 295
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 3.2,
|
393 |
+
"learning_rate": 0.00046012269938650307,
|
394 |
+
"loss": 1.1005,
|
395 |
+
"step": 300
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 3.25,
|
399 |
+
"learning_rate": 0.0004677914110429448,
|
400 |
+
"loss": 1.104,
|
401 |
+
"step": 305
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 3.31,
|
405 |
+
"learning_rate": 0.0004754601226993865,
|
406 |
+
"loss": 1.1068,
|
407 |
+
"step": 310
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 3.36,
|
411 |
+
"learning_rate": 0.00048312883435582823,
|
412 |
+
"loss": 1.0989,
|
413 |
+
"step": 315
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 3.41,
|
417 |
+
"learning_rate": 0.00049079754601227,
|
418 |
+
"loss": 1.0993,
|
419 |
+
"step": 320
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 3.47,
|
423 |
+
"learning_rate": 0.0004984662576687117,
|
424 |
+
"loss": 1.099,
|
425 |
+
"step": 325
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 3.52,
|
429 |
+
"learning_rate": 0.0004938461538461538,
|
430 |
+
"loss": 1.0975,
|
431 |
+
"step": 330
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 3.57,
|
435 |
+
"learning_rate": 0.0004861538461538462,
|
436 |
+
"loss": 1.1046,
|
437 |
+
"step": 335
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 3.63,
|
441 |
+
"learning_rate": 0.00047846153846153844,
|
442 |
+
"loss": 1.1051,
|
443 |
+
"step": 340
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 3.68,
|
447 |
+
"learning_rate": 0.00047076923076923077,
|
448 |
+
"loss": 1.096,
|
449 |
+
"step": 345
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 3.73,
|
453 |
+
"learning_rate": 0.00046307692307692304,
|
454 |
+
"loss": 1.104,
|
455 |
+
"step": 350
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 3.79,
|
459 |
+
"learning_rate": 0.0004553846153846154,
|
460 |
+
"loss": 1.1024,
|
461 |
+
"step": 355
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 3.84,
|
465 |
+
"learning_rate": 0.0004476923076923077,
|
466 |
+
"loss": 1.1037,
|
467 |
+
"step": 360
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 3.89,
|
471 |
+
"learning_rate": 0.00044,
|
472 |
+
"loss": 1.1012,
|
473 |
+
"step": 365
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 3.95,
|
477 |
+
"learning_rate": 0.0004323076923076923,
|
478 |
+
"loss": 1.0989,
|
479 |
+
"step": 370
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 4.0,
|
483 |
+
"learning_rate": 0.0004246153846153846,
|
484 |
+
"loss": 1.0934,
|
485 |
+
"step": 375
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 4.0,
|
489 |
+
"eval_accuracy": 0.6431893687707642,
|
490 |
+
"eval_loss": 1.0786629915237427,
|
491 |
+
"eval_runtime": 149.1448,
|
492 |
+
"eval_samples_per_second": 30.273,
|
493 |
+
"eval_steps_per_second": 1.897,
|
494 |
+
"step": 375
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 4.05,
|
498 |
+
"learning_rate": 0.0004169230769230769,
|
499 |
+
"loss": 1.0464,
|
500 |
+
"step": 380
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 4.11,
|
504 |
+
"learning_rate": 0.00040923076923076926,
|
505 |
+
"loss": 0.729,
|
506 |
+
"step": 385
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 4.16,
|
510 |
+
"learning_rate": 0.00040153846153846153,
|
511 |
+
"loss": 0.6866,
|
512 |
+
"step": 390
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 4.21,
|
516 |
+
"learning_rate": 0.00039384615384615386,
|
517 |
+
"loss": 0.6134,
|
518 |
+
"step": 395
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 4.27,
|
522 |
+
"learning_rate": 0.0003861538461538462,
|
523 |
+
"loss": 0.5426,
|
524 |
+
"step": 400
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 4.32,
|
528 |
+
"learning_rate": 0.00037846153846153845,
|
529 |
+
"loss": 0.5194,
|
530 |
+
"step": 405
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 4.37,
|
534 |
+
"learning_rate": 0.0003707692307692308,
|
535 |
+
"loss": 0.598,
|
536 |
+
"step": 410
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 4.43,
|
540 |
+
"learning_rate": 0.0003630769230769231,
|
541 |
+
"loss": 0.6058,
|
542 |
+
"step": 415
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 4.48,
|
546 |
+
"learning_rate": 0.00035538461538461543,
|
547 |
+
"loss": 0.5753,
|
548 |
+
"step": 420
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 4.53,
|
552 |
+
"learning_rate": 0.0003476923076923077,
|
553 |
+
"loss": 0.619,
|
554 |
+
"step": 425
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 4.59,
|
558 |
+
"learning_rate": 0.00034,
|
559 |
+
"loss": 0.8952,
|
560 |
+
"step": 430
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 4.64,
|
564 |
+
"learning_rate": 0.0003323076923076923,
|
565 |
+
"loss": 0.619,
|
566 |
+
"step": 435
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 4.69,
|
570 |
+
"learning_rate": 0.0003246153846153846,
|
571 |
+
"loss": 0.5832,
|
572 |
+
"step": 440
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 4.75,
|
576 |
+
"learning_rate": 0.00031692307692307695,
|
577 |
+
"loss": 0.5946,
|
578 |
+
"step": 445
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 4.8,
|
582 |
+
"learning_rate": 0.00030923076923076927,
|
583 |
+
"loss": 0.5846,
|
584 |
+
"step": 450
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 4.85,
|
588 |
+
"learning_rate": 0.00030153846153846154,
|
589 |
+
"loss": 0.5214,
|
590 |
+
"step": 455
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 4.91,
|
594 |
+
"learning_rate": 0.00029384615384615387,
|
595 |
+
"loss": 0.4885,
|
596 |
+
"step": 460
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 4.96,
|
600 |
+
"learning_rate": 0.00028615384615384614,
|
601 |
+
"loss": 0.5795,
|
602 |
+
"step": 465
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 4.99,
|
606 |
+
"eval_accuracy": 0.6445182724252492,
|
607 |
+
"eval_loss": 0.5864280462265015,
|
608 |
+
"eval_runtime": 148.9609,
|
609 |
+
"eval_samples_per_second": 30.31,
|
610 |
+
"eval_steps_per_second": 1.9,
|
611 |
+
"step": 468
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 5.01,
|
615 |
+
"learning_rate": 0.00027846153846153846,
|
616 |
+
"loss": 0.5634,
|
617 |
+
"step": 470
|
618 |
+
},
|
619 |
+
{
|
620 |
+
"epoch": 5.07,
|
621 |
+
"learning_rate": 0.0002707692307692308,
|
622 |
+
"loss": 0.639,
|
623 |
+
"step": 475
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 5.12,
|
627 |
+
"learning_rate": 0.0002630769230769231,
|
628 |
+
"loss": 0.5783,
|
629 |
+
"step": 480
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 5.17,
|
633 |
+
"learning_rate": 0.0002553846153846154,
|
634 |
+
"loss": 0.5436,
|
635 |
+
"step": 485
|
636 |
+
},
|
637 |
+
{
|
638 |
+
"epoch": 5.23,
|
639 |
+
"learning_rate": 0.0002476923076923077,
|
640 |
+
"loss": 0.547,
|
641 |
+
"step": 490
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 5.28,
|
645 |
+
"learning_rate": 0.00024,
|
646 |
+
"loss": 0.6463,
|
647 |
+
"step": 495
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 5.33,
|
651 |
+
"learning_rate": 0.0002323076923076923,
|
652 |
+
"loss": 0.5136,
|
653 |
+
"step": 500
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 5.39,
|
657 |
+
"learning_rate": 0.0002246153846153846,
|
658 |
+
"loss": 0.6011,
|
659 |
+
"step": 505
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 5.44,
|
663 |
+
"learning_rate": 0.00021692307692307693,
|
664 |
+
"loss": 0.5484,
|
665 |
+
"step": 510
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 5.49,
|
669 |
+
"learning_rate": 0.00020923076923076922,
|
670 |
+
"loss": 0.5498,
|
671 |
+
"step": 515
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 5.55,
|
675 |
+
"learning_rate": 0.00020153846153846152,
|
676 |
+
"loss": 0.5498,
|
677 |
+
"step": 520
|
678 |
+
},
|
679 |
+
{
|
680 |
+
"epoch": 5.6,
|
681 |
+
"learning_rate": 0.00019384615384615385,
|
682 |
+
"loss": 0.5326,
|
683 |
+
"step": 525
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 5.65,
|
687 |
+
"learning_rate": 0.00018615384615384615,
|
688 |
+
"loss": 0.548,
|
689 |
+
"step": 530
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 5.71,
|
693 |
+
"learning_rate": 0.00017846153846153844,
|
694 |
+
"loss": 0.5529,
|
695 |
+
"step": 535
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 5.76,
|
699 |
+
"learning_rate": 0.00017076923076923077,
|
700 |
+
"loss": 0.5377,
|
701 |
+
"step": 540
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 5.81,
|
705 |
+
"learning_rate": 0.00016307692307692307,
|
706 |
+
"loss": 0.5506,
|
707 |
+
"step": 545
|
708 |
+
},
|
709 |
+
{
|
710 |
+
"epoch": 5.87,
|
711 |
+
"learning_rate": 0.00015538461538461536,
|
712 |
+
"loss": 0.5779,
|
713 |
+
"step": 550
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 5.92,
|
717 |
+
"learning_rate": 0.00014769230769230772,
|
718 |
+
"loss": 0.5705,
|
719 |
+
"step": 555
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 5.97,
|
723 |
+
"learning_rate": 0.00014000000000000001,
|
724 |
+
"loss": 0.5437,
|
725 |
+
"step": 560
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 5.99,
|
729 |
+
"eval_accuracy": 0.7368770764119601,
|
730 |
+
"eval_loss": 0.5732536315917969,
|
731 |
+
"eval_runtime": 149.4798,
|
732 |
+
"eval_samples_per_second": 30.205,
|
733 |
+
"eval_steps_per_second": 1.893,
|
734 |
+
"step": 562
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 6.03,
|
738 |
+
"learning_rate": 0.0001323076923076923,
|
739 |
+
"loss": 0.5752,
|
740 |
+
"step": 565
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 6.08,
|
744 |
+
"learning_rate": 0.0001246153846153846,
|
745 |
+
"loss": 0.5415,
|
746 |
+
"step": 570
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 6.13,
|
750 |
+
"learning_rate": 0.00011692307692307692,
|
751 |
+
"loss": 0.5068,
|
752 |
+
"step": 575
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 6.19,
|
756 |
+
"learning_rate": 0.00010923076923076923,
|
757 |
+
"loss": 0.5202,
|
758 |
+
"step": 580
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 6.24,
|
762 |
+
"learning_rate": 0.00010153846153846153,
|
763 |
+
"loss": 0.4475,
|
764 |
+
"step": 585
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 6.29,
|
768 |
+
"learning_rate": 9.384615384615384e-05,
|
769 |
+
"loss": 0.4511,
|
770 |
+
"step": 590
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 6.35,
|
774 |
+
"learning_rate": 8.615384615384615e-05,
|
775 |
+
"loss": 0.4405,
|
776 |
+
"step": 595
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 6.4,
|
780 |
+
"learning_rate": 7.846153846153845e-05,
|
781 |
+
"loss": 0.384,
|
782 |
+
"step": 600
|
783 |
+
},
|
784 |
+
{
|
785 |
+
"epoch": 6.45,
|
786 |
+
"learning_rate": 7.076923076923076e-05,
|
787 |
+
"loss": 0.3895,
|
788 |
+
"step": 605
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 6.51,
|
792 |
+
"learning_rate": 6.307692307692308e-05,
|
793 |
+
"loss": 0.3673,
|
794 |
+
"step": 610
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 6.56,
|
798 |
+
"learning_rate": 5.538461538461539e-05,
|
799 |
+
"loss": 0.3665,
|
800 |
+
"step": 615
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 6.61,
|
804 |
+
"learning_rate": 4.769230769230769e-05,
|
805 |
+
"loss": 0.3448,
|
806 |
+
"step": 620
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 6.67,
|
810 |
+
"learning_rate": 4e-05,
|
811 |
+
"loss": 0.3582,
|
812 |
+
"step": 625
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 6.72,
|
816 |
+
"learning_rate": 3.230769230769231e-05,
|
817 |
+
"loss": 0.3487,
|
818 |
+
"step": 630
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 6.77,
|
822 |
+
"learning_rate": 2.4615384615384616e-05,
|
823 |
+
"loss": 0.3195,
|
824 |
+
"step": 635
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 6.83,
|
828 |
+
"learning_rate": 1.6923076923076924e-05,
|
829 |
+
"loss": 0.3806,
|
830 |
+
"step": 640
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 6.88,
|
834 |
+
"learning_rate": 9.230769230769232e-06,
|
835 |
+
"loss": 0.3676,
|
836 |
+
"step": 645
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 6.93,
|
840 |
+
"learning_rate": 1.5384615384615385e-06,
|
841 |
+
"loss": 0.3369,
|
842 |
+
"step": 650
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 6.94,
|
846 |
+
"eval_accuracy": 0.9029900332225914,
|
847 |
+
"eval_loss": 0.32981640100479126,
|
848 |
+
"eval_runtime": 149.9457,
|
849 |
+
"eval_samples_per_second": 30.111,
|
850 |
+
"eval_steps_per_second": 1.887,
|
851 |
+
"step": 651
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 6.94,
|
855 |
+
"step": 651,
|
856 |
+
"total_flos": 7.375405476885369e+18,
|
857 |
+
"train_loss": 0.5960829522569425,
|
858 |
+
"train_runtime": 11023.3494,
|
859 |
+
"train_samples_per_second": 3.801,
|
860 |
+
"train_steps_per_second": 0.059
|
861 |
+
}
|
862 |
+
],
|
863 |
+
"max_steps": 651,
|
864 |
+
"num_train_epochs": 7,
|
865 |
+
"total_flos": 7.375405476885369e+18,
|
866 |
+
"trial_name": null,
|
867 |
+
"trial_params": null
|
868 |
+
}
|