AlekseyKorshuk
commited on
Commit
•
36de6eb
1
Parent(s):
f56333f
End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +211 -0
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 2.5182421875,
|
4 |
+
"train_runtime": 15501.0618,
|
5 |
+
"train_samples": 599,
|
6 |
+
"train_samples_per_second": 0.039,
|
7 |
+
"train_steps_per_second": 0.005
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 2.5182421875,
|
4 |
+
"train_runtime": 15501.0618,
|
5 |
+
"train_samples": 599,
|
6 |
+
"train_samples_per_second": 0.039,
|
7 |
+
"train_steps_per_second": 0.005
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,211 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"global_step": 75,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 1e-06,
|
13 |
+
"loss": 2.718,
|
14 |
+
"step": 1
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.08,
|
18 |
+
"learning_rate": 1e-06,
|
19 |
+
"loss": 2.5852,
|
20 |
+
"step": 6
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.08,
|
24 |
+
"eval_accuracy": 0.26974484268246846,
|
25 |
+
"eval_loss": 2.595703125,
|
26 |
+
"eval_runtime": 73.3015,
|
27 |
+
"eval_samples_per_second": 4.966,
|
28 |
+
"eval_steps_per_second": 0.628,
|
29 |
+
"step": 6
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.16,
|
33 |
+
"learning_rate": 1e-06,
|
34 |
+
"loss": 2.5956,
|
35 |
+
"step": 12
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 0.16,
|
39 |
+
"eval_accuracy": 0.27058504875406286,
|
40 |
+
"eval_loss": 2.576171875,
|
41 |
+
"eval_runtime": 73.9801,
|
42 |
+
"eval_samples_per_second": 4.92,
|
43 |
+
"eval_steps_per_second": 0.622,
|
44 |
+
"step": 12
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.24,
|
48 |
+
"learning_rate": 1e-06,
|
49 |
+
"loss": 2.5961,
|
50 |
+
"step": 18
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.24,
|
54 |
+
"eval_accuracy": 0.27107148384814383,
|
55 |
+
"eval_loss": 2.5546875,
|
56 |
+
"eval_runtime": 73.8954,
|
57 |
+
"eval_samples_per_second": 4.926,
|
58 |
+
"eval_steps_per_second": 0.623,
|
59 |
+
"step": 18
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 0.32,
|
63 |
+
"learning_rate": 1e-06,
|
64 |
+
"loss": 2.5731,
|
65 |
+
"step": 24
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.32,
|
69 |
+
"eval_accuracy": 0.27218807349592056,
|
70 |
+
"eval_loss": 2.53125,
|
71 |
+
"eval_runtime": 73.9168,
|
72 |
+
"eval_samples_per_second": 4.924,
|
73 |
+
"eval_steps_per_second": 0.622,
|
74 |
+
"step": 24
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.4,
|
78 |
+
"learning_rate": 1e-06,
|
79 |
+
"loss": 2.5415,
|
80 |
+
"step": 30
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.4,
|
84 |
+
"eval_accuracy": 0.2733599398589338,
|
85 |
+
"eval_loss": 2.51171875,
|
86 |
+
"eval_runtime": 73.9155,
|
87 |
+
"eval_samples_per_second": 4.925,
|
88 |
+
"eval_steps_per_second": 0.622,
|
89 |
+
"step": 30
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 0.48,
|
93 |
+
"learning_rate": 1e-06,
|
94 |
+
"loss": 2.5168,
|
95 |
+
"step": 36
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 0.48,
|
99 |
+
"eval_accuracy": 0.2745704999226126,
|
100 |
+
"eval_loss": 2.49609375,
|
101 |
+
"eval_runtime": 73.8737,
|
102 |
+
"eval_samples_per_second": 4.927,
|
103 |
+
"eval_steps_per_second": 0.623,
|
104 |
+
"step": 36
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.56,
|
108 |
+
"learning_rate": 1e-06,
|
109 |
+
"loss": 2.4972,
|
110 |
+
"step": 42
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.56,
|
114 |
+
"eval_accuracy": 0.2755544254538218,
|
115 |
+
"eval_loss": 2.482421875,
|
116 |
+
"eval_runtime": 73.9197,
|
117 |
+
"eval_samples_per_second": 4.924,
|
118 |
+
"eval_steps_per_second": 0.622,
|
119 |
+
"step": 42
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 0.64,
|
123 |
+
"learning_rate": 1e-06,
|
124 |
+
"loss": 2.4354,
|
125 |
+
"step": 48
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 0.64,
|
129 |
+
"eval_accuracy": 0.27611824794923384,
|
130 |
+
"eval_loss": 2.47265625,
|
131 |
+
"eval_runtime": 74.1261,
|
132 |
+
"eval_samples_per_second": 4.911,
|
133 |
+
"eval_steps_per_second": 0.621,
|
134 |
+
"step": 48
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.72,
|
138 |
+
"learning_rate": 1e-06,
|
139 |
+
"loss": 2.4055,
|
140 |
+
"step": 54
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.72,
|
144 |
+
"eval_accuracy": 0.2768423729188317,
|
145 |
+
"eval_loss": 2.4609375,
|
146 |
+
"eval_runtime": 73.8501,
|
147 |
+
"eval_samples_per_second": 4.929,
|
148 |
+
"eval_steps_per_second": 0.623,
|
149 |
+
"step": 54
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.8,
|
153 |
+
"learning_rate": 1e-06,
|
154 |
+
"loss": 2.4681,
|
155 |
+
"step": 60
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 0.8,
|
159 |
+
"eval_accuracy": 0.27778207707785174,
|
160 |
+
"eval_loss": 2.44921875,
|
161 |
+
"eval_runtime": 73.8512,
|
162 |
+
"eval_samples_per_second": 4.929,
|
163 |
+
"eval_steps_per_second": 0.623,
|
164 |
+
"step": 60
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.88,
|
168 |
+
"learning_rate": 1e-06,
|
169 |
+
"loss": 2.5866,
|
170 |
+
"step": 66
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.88,
|
174 |
+
"eval_accuracy": 0.27841223163154755,
|
175 |
+
"eval_loss": 2.435546875,
|
176 |
+
"eval_runtime": 73.8943,
|
177 |
+
"eval_samples_per_second": 4.926,
|
178 |
+
"eval_steps_per_second": 0.623,
|
179 |
+
"step": 66
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 0.96,
|
183 |
+
"learning_rate": 1e-06,
|
184 |
+
"loss": 2.4221,
|
185 |
+
"step": 72
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 0.96,
|
189 |
+
"eval_accuracy": 0.27929113140380746,
|
190 |
+
"eval_loss": 2.423828125,
|
191 |
+
"eval_runtime": 73.8857,
|
192 |
+
"eval_samples_per_second": 4.927,
|
193 |
+
"eval_steps_per_second": 0.623,
|
194 |
+
"step": 72
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 1.0,
|
198 |
+
"step": 75,
|
199 |
+
"total_flos": 4974649540608.0,
|
200 |
+
"train_loss": 2.5182421875,
|
201 |
+
"train_runtime": 15501.0618,
|
202 |
+
"train_samples_per_second": 0.039,
|
203 |
+
"train_steps_per_second": 0.005
|
204 |
+
}
|
205 |
+
],
|
206 |
+
"max_steps": 75,
|
207 |
+
"num_train_epochs": 1,
|
208 |
+
"total_flos": 4974649540608.0,
|
209 |
+
"trial_name": null,
|
210 |
+
"trial_params": null
|
211 |
+
}
|