Commit
•
3d40c2f
1
Parent(s):
3a6fc78
Upload trainer_state.json
Browse files- trainer_state.json +577 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,577 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.015090291388332844,
|
| 3 |
+
"best_model_checkpoint": "./convnext-tiny-mask-finetuned/checkpoint-500",
|
| 4 |
+
"epoch": 3.5398230088495577,
|
| 5 |
+
"global_step": 800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"learning_rate": 0.00019778761061946905,
|
| 13 |
+
"loss": 0.0126,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.09,
|
| 18 |
+
"learning_rate": 0.00019557522123893806,
|
| 19 |
+
"loss": 0.0237,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.13,
|
| 24 |
+
"learning_rate": 0.0001933628318584071,
|
| 25 |
+
"loss": 0.0026,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.18,
|
| 30 |
+
"learning_rate": 0.00019115044247787613,
|
| 31 |
+
"loss": 0.0213,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.22,
|
| 36 |
+
"learning_rate": 0.00018893805309734514,
|
| 37 |
+
"loss": 0.01,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.27,
|
| 42 |
+
"learning_rate": 0.00018672566371681417,
|
| 43 |
+
"loss": 0.0325,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.31,
|
| 48 |
+
"learning_rate": 0.0001845132743362832,
|
| 49 |
+
"loss": 0.0212,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.35,
|
| 54 |
+
"learning_rate": 0.00018230088495575222,
|
| 55 |
+
"loss": 0.0138,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.4,
|
| 60 |
+
"learning_rate": 0.00018008849557522126,
|
| 61 |
+
"loss": 0.0175,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.44,
|
| 66 |
+
"learning_rate": 0.0001778761061946903,
|
| 67 |
+
"loss": 0.013,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.44,
|
| 72 |
+
"eval_accuracy": 0.9944382647385984,
|
| 73 |
+
"eval_loss": 0.023499028757214546,
|
| 74 |
+
"eval_runtime": 80.0912,
|
| 75 |
+
"eval_samples_per_second": 44.899,
|
| 76 |
+
"eval_steps_per_second": 5.619,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.49,
|
| 81 |
+
"learning_rate": 0.0001756637168141593,
|
| 82 |
+
"loss": 0.0188,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.53,
|
| 87 |
+
"learning_rate": 0.00017345132743362834,
|
| 88 |
+
"loss": 0.0157,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.58,
|
| 93 |
+
"learning_rate": 0.00017123893805309737,
|
| 94 |
+
"loss": 0.0108,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.62,
|
| 99 |
+
"learning_rate": 0.00016902654867256638,
|
| 100 |
+
"loss": 0.0125,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.66,
|
| 105 |
+
"learning_rate": 0.00016681415929203542,
|
| 106 |
+
"loss": 0.0165,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.71,
|
| 111 |
+
"learning_rate": 0.00016460176991150443,
|
| 112 |
+
"loss": 0.0226,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.75,
|
| 117 |
+
"learning_rate": 0.00016238938053097346,
|
| 118 |
+
"loss": 0.0083,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.8,
|
| 123 |
+
"learning_rate": 0.0001601769911504425,
|
| 124 |
+
"loss": 0.0231,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.84,
|
| 129 |
+
"learning_rate": 0.0001579646017699115,
|
| 130 |
+
"loss": 0.0151,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.88,
|
| 135 |
+
"learning_rate": 0.00015575221238938055,
|
| 136 |
+
"loss": 0.014,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.88,
|
| 141 |
+
"eval_accuracy": 0.9936040044493882,
|
| 142 |
+
"eval_loss": 0.01512084435671568,
|
| 143 |
+
"eval_runtime": 80.2971,
|
| 144 |
+
"eval_samples_per_second": 44.784,
|
| 145 |
+
"eval_steps_per_second": 5.604,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.93,
|
| 150 |
+
"learning_rate": 0.00015353982300884958,
|
| 151 |
+
"loss": 0.0195,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.97,
|
| 156 |
+
"learning_rate": 0.0001513274336283186,
|
| 157 |
+
"loss": 0.0328,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.02,
|
| 162 |
+
"learning_rate": 0.00014911504424778763,
|
| 163 |
+
"loss": 0.0053,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.06,
|
| 168 |
+
"learning_rate": 0.00014690265486725664,
|
| 169 |
+
"loss": 0.004,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.11,
|
| 174 |
+
"learning_rate": 0.00014469026548672567,
|
| 175 |
+
"loss": 0.021,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.15,
|
| 180 |
+
"learning_rate": 0.00014247787610619468,
|
| 181 |
+
"loss": 0.0078,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.19,
|
| 186 |
+
"learning_rate": 0.00014026548672566372,
|
| 187 |
+
"loss": 0.0149,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.24,
|
| 192 |
+
"learning_rate": 0.00013805309734513276,
|
| 193 |
+
"loss": 0.0149,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.28,
|
| 198 |
+
"learning_rate": 0.00013584070796460177,
|
| 199 |
+
"loss": 0.0052,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.33,
|
| 204 |
+
"learning_rate": 0.0001336283185840708,
|
| 205 |
+
"loss": 0.0022,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.33,
|
| 210 |
+
"eval_accuracy": 0.9947163515016685,
|
| 211 |
+
"eval_loss": 0.017131101340055466,
|
| 212 |
+
"eval_runtime": 80.9561,
|
| 213 |
+
"eval_samples_per_second": 44.419,
|
| 214 |
+
"eval_steps_per_second": 5.559,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.37,
|
| 219 |
+
"learning_rate": 0.0001314159292035398,
|
| 220 |
+
"loss": 0.0056,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.42,
|
| 225 |
+
"learning_rate": 0.00012920353982300885,
|
| 226 |
+
"loss": 0.0176,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.46,
|
| 231 |
+
"learning_rate": 0.00012699115044247788,
|
| 232 |
+
"loss": 0.0066,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.5,
|
| 237 |
+
"learning_rate": 0.0001247787610619469,
|
| 238 |
+
"loss": 0.0139,
|
| 239 |
+
"step": 340
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.55,
|
| 243 |
+
"learning_rate": 0.00012256637168141593,
|
| 244 |
+
"loss": 0.0056,
|
| 245 |
+
"step": 350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.59,
|
| 249 |
+
"learning_rate": 0.00012035398230088497,
|
| 250 |
+
"loss": 0.0009,
|
| 251 |
+
"step": 360
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 1.64,
|
| 255 |
+
"learning_rate": 0.00011814159292035399,
|
| 256 |
+
"loss": 0.0004,
|
| 257 |
+
"step": 370
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.68,
|
| 261 |
+
"learning_rate": 0.00011592920353982301,
|
| 262 |
+
"loss": 0.0094,
|
| 263 |
+
"step": 380
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.73,
|
| 267 |
+
"learning_rate": 0.00011371681415929205,
|
| 268 |
+
"loss": 0.0037,
|
| 269 |
+
"step": 390
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.77,
|
| 273 |
+
"learning_rate": 0.00011150442477876106,
|
| 274 |
+
"loss": 0.0022,
|
| 275 |
+
"step": 400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.77,
|
| 279 |
+
"eval_accuracy": 0.9958286985539488,
|
| 280 |
+
"eval_loss": 0.01814076118171215,
|
| 281 |
+
"eval_runtime": 80.2281,
|
| 282 |
+
"eval_samples_per_second": 44.822,
|
| 283 |
+
"eval_steps_per_second": 5.609,
|
| 284 |
+
"step": 400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.81,
|
| 288 |
+
"learning_rate": 0.00010929203539823009,
|
| 289 |
+
"loss": 0.01,
|
| 290 |
+
"step": 410
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.86,
|
| 294 |
+
"learning_rate": 0.00010707964601769913,
|
| 295 |
+
"loss": 0.0117,
|
| 296 |
+
"step": 420
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.9,
|
| 300 |
+
"learning_rate": 0.00010486725663716814,
|
| 301 |
+
"loss": 0.0055,
|
| 302 |
+
"step": 430
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.95,
|
| 306 |
+
"learning_rate": 0.00010265486725663717,
|
| 307 |
+
"loss": 0.0069,
|
| 308 |
+
"step": 440
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.99,
|
| 312 |
+
"learning_rate": 0.00010044247787610618,
|
| 313 |
+
"loss": 0.0102,
|
| 314 |
+
"step": 450
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.04,
|
| 318 |
+
"learning_rate": 9.823008849557522e-05,
|
| 319 |
+
"loss": 0.0018,
|
| 320 |
+
"step": 460
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.08,
|
| 324 |
+
"learning_rate": 9.601769911504426e-05,
|
| 325 |
+
"loss": 0.0027,
|
| 326 |
+
"step": 470
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.12,
|
| 330 |
+
"learning_rate": 9.380530973451328e-05,
|
| 331 |
+
"loss": 0.0115,
|
| 332 |
+
"step": 480
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 2.17,
|
| 336 |
+
"learning_rate": 9.15929203539823e-05,
|
| 337 |
+
"loss": 0.0098,
|
| 338 |
+
"step": 490
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.21,
|
| 342 |
+
"learning_rate": 8.938053097345133e-05,
|
| 343 |
+
"loss": 0.0037,
|
| 344 |
+
"step": 500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.21,
|
| 348 |
+
"eval_accuracy": 0.9961067853170189,
|
| 349 |
+
"eval_loss": 0.015090291388332844,
|
| 350 |
+
"eval_runtime": 80.1059,
|
| 351 |
+
"eval_samples_per_second": 44.891,
|
| 352 |
+
"eval_steps_per_second": 5.618,
|
| 353 |
+
"step": 500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 2.26,
|
| 357 |
+
"learning_rate": 8.716814159292036e-05,
|
| 358 |
+
"loss": 0.0031,
|
| 359 |
+
"step": 510
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.3,
|
| 363 |
+
"learning_rate": 8.495575221238938e-05,
|
| 364 |
+
"loss": 0.0033,
|
| 365 |
+
"step": 520
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 2.35,
|
| 369 |
+
"learning_rate": 8.274336283185841e-05,
|
| 370 |
+
"loss": 0.0005,
|
| 371 |
+
"step": 530
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 2.39,
|
| 375 |
+
"learning_rate": 8.053097345132744e-05,
|
| 376 |
+
"loss": 0.0008,
|
| 377 |
+
"step": 540
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.43,
|
| 381 |
+
"learning_rate": 7.831858407079647e-05,
|
| 382 |
+
"loss": 0.0008,
|
| 383 |
+
"step": 550
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 2.48,
|
| 387 |
+
"learning_rate": 7.610619469026549e-05,
|
| 388 |
+
"loss": 0.0009,
|
| 389 |
+
"step": 560
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.52,
|
| 393 |
+
"learning_rate": 7.389380530973453e-05,
|
| 394 |
+
"loss": 0.0001,
|
| 395 |
+
"step": 570
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 2.57,
|
| 399 |
+
"learning_rate": 7.168141592920355e-05,
|
| 400 |
+
"loss": 0.0004,
|
| 401 |
+
"step": 580
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 2.61,
|
| 405 |
+
"learning_rate": 6.946902654867257e-05,
|
| 406 |
+
"loss": 0.0084,
|
| 407 |
+
"step": 590
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 2.65,
|
| 411 |
+
"learning_rate": 6.725663716814161e-05,
|
| 412 |
+
"loss": 0.0102,
|
| 413 |
+
"step": 600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.65,
|
| 417 |
+
"eval_accuracy": 0.9949944382647385,
|
| 418 |
+
"eval_loss": 0.016862528398633003,
|
| 419 |
+
"eval_runtime": 79.8413,
|
| 420 |
+
"eval_samples_per_second": 45.039,
|
| 421 |
+
"eval_steps_per_second": 5.636,
|
| 422 |
+
"step": 600
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.7,
|
| 426 |
+
"learning_rate": 6.504424778761063e-05,
|
| 427 |
+
"loss": 0.0002,
|
| 428 |
+
"step": 610
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.74,
|
| 432 |
+
"learning_rate": 6.283185840707965e-05,
|
| 433 |
+
"loss": 0.0035,
|
| 434 |
+
"step": 620
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.79,
|
| 438 |
+
"learning_rate": 6.061946902654868e-05,
|
| 439 |
+
"loss": 0.0017,
|
| 440 |
+
"step": 630
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 2.83,
|
| 444 |
+
"learning_rate": 5.8407079646017705e-05,
|
| 445 |
+
"loss": 0.0004,
|
| 446 |
+
"step": 640
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.88,
|
| 450 |
+
"learning_rate": 5.619469026548673e-05,
|
| 451 |
+
"loss": 0.0001,
|
| 452 |
+
"step": 650
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.92,
|
| 456 |
+
"learning_rate": 5.398230088495575e-05,
|
| 457 |
+
"loss": 0.0,
|
| 458 |
+
"step": 660
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 2.96,
|
| 462 |
+
"learning_rate": 5.176991150442479e-05,
|
| 463 |
+
"loss": 0.0003,
|
| 464 |
+
"step": 670
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.01,
|
| 468 |
+
"learning_rate": 4.955752212389381e-05,
|
| 469 |
+
"loss": 0.0005,
|
| 470 |
+
"step": 680
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 3.05,
|
| 474 |
+
"learning_rate": 4.734513274336283e-05,
|
| 475 |
+
"loss": 0.0001,
|
| 476 |
+
"step": 690
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 3.1,
|
| 480 |
+
"learning_rate": 4.5132743362831855e-05,
|
| 481 |
+
"loss": 0.0003,
|
| 482 |
+
"step": 700
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 3.1,
|
| 486 |
+
"eval_accuracy": 0.9961067853170189,
|
| 487 |
+
"eval_loss": 0.01964888535439968,
|
| 488 |
+
"eval_runtime": 79.8617,
|
| 489 |
+
"eval_samples_per_second": 45.028,
|
| 490 |
+
"eval_steps_per_second": 5.635,
|
| 491 |
+
"step": 700
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.14,
|
| 495 |
+
"learning_rate": 4.2920353982300885e-05,
|
| 496 |
+
"loss": 0.0001,
|
| 497 |
+
"step": 710
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 3.19,
|
| 501 |
+
"learning_rate": 4.0707964601769914e-05,
|
| 502 |
+
"loss": 0.0007,
|
| 503 |
+
"step": 720
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 3.23,
|
| 507 |
+
"learning_rate": 3.849557522123894e-05,
|
| 508 |
+
"loss": 0.0001,
|
| 509 |
+
"step": 730
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 3.27,
|
| 513 |
+
"learning_rate": 3.628318584070797e-05,
|
| 514 |
+
"loss": 0.0017,
|
| 515 |
+
"step": 740
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 3.32,
|
| 519 |
+
"learning_rate": 3.407079646017699e-05,
|
| 520 |
+
"loss": 0.0006,
|
| 521 |
+
"step": 750
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 3.36,
|
| 525 |
+
"learning_rate": 3.185840707964602e-05,
|
| 526 |
+
"loss": 0.0003,
|
| 527 |
+
"step": 760
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 3.41,
|
| 531 |
+
"learning_rate": 2.964601769911505e-05,
|
| 532 |
+
"loss": 0.0001,
|
| 533 |
+
"step": 770
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 3.45,
|
| 537 |
+
"learning_rate": 2.743362831858407e-05,
|
| 538 |
+
"loss": 0.0001,
|
| 539 |
+
"step": 780
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 3.5,
|
| 543 |
+
"learning_rate": 2.5221238938053098e-05,
|
| 544 |
+
"loss": 0.0008,
|
| 545 |
+
"step": 790
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 3.54,
|
| 549 |
+
"learning_rate": 2.3008849557522124e-05,
|
| 550 |
+
"loss": 0.0002,
|
| 551 |
+
"step": 800
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 3.54,
|
| 555 |
+
"eval_accuracy": 0.9955506117908788,
|
| 556 |
+
"eval_loss": 0.017847377806901932,
|
| 557 |
+
"eval_runtime": 81.2034,
|
| 558 |
+
"eval_samples_per_second": 44.284,
|
| 559 |
+
"eval_steps_per_second": 5.542,
|
| 560 |
+
"step": 800
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 3.54,
|
| 564 |
+
"step": 800,
|
| 565 |
+
"total_flos": 1.2838213030685737e+18,
|
| 566 |
+
"train_loss": 0.007916164267808199,
|
| 567 |
+
"train_runtime": 4100.2539,
|
| 568 |
+
"train_samples_per_second": 14.075,
|
| 569 |
+
"train_steps_per_second": 0.22
|
| 570 |
+
}
|
| 571 |
+
],
|
| 572 |
+
"max_steps": 904,
|
| 573 |
+
"num_train_epochs": 4,
|
| 574 |
+
"total_flos": 1.2838213030685737e+18,
|
| 575 |
+
"trial_name": null,
|
| 576 |
+
"trial_params": null
|
| 577 |
+
}
|