Training in progress, epoch 1
Browse files- all_results.json +11 -11
- eval_results.json +6 -6
- model.safetensors +1 -1
- runs/Nov18_12-36-05_fac6bf2076b2/events.out.tfevents.1731936395.fac6bf2076b2.31.2 +3 -0
- runs/Nov18_13-26-46_fac6bf2076b2/events.out.tfevents.1731936413.fac6bf2076b2.31.3 +3 -0
- train_results.json +6 -6
- trainer_state.json +376 -263
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 5.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8257839721254355,
|
| 4 |
+
"eval_loss": 0.4613528847694397,
|
| 5 |
+
"eval_runtime": 5.98,
|
| 6 |
+
"eval_samples_per_second": 143.98,
|
| 7 |
+
"eval_steps_per_second": 2.341,
|
| 8 |
+
"total_flos": 4.064614037073838e+18,
|
| 9 |
+
"train_loss": 0.8882290616631507,
|
| 10 |
+
"train_runtime": 1944.8842,
|
| 11 |
+
"train_samples_per_second": 84.062,
|
| 12 |
+
"train_steps_per_second": 0.329
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 5.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8257839721254355,
|
| 4 |
+
"eval_loss": 0.4613528847694397,
|
| 5 |
+
"eval_runtime": 5.98,
|
| 6 |
+
"eval_samples_per_second": 143.98,
|
| 7 |
+
"eval_steps_per_second": 2.341
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 110367448
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9816ed93b4eec04bf23e2baa39a074364eb3de0c2e6133c698f1a52544d9b6f7
|
| 3 |
size 110367448
|
runs/Nov18_12-36-05_fac6bf2076b2/events.out.tfevents.1731936395.fac6bf2076b2.31.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:654bcb90f6650830fd65aa4a2217c94bbb71c57e2dd1c5d3c90a646e94a1dd88
|
| 3 |
+
size 411
|
runs/Nov18_13-26-46_fac6bf2076b2/events.out.tfevents.1731936413.fac6bf2076b2.31.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4915d5ee1f7f08b36ecd656375c4e53252763f404642e3c3cb96d9406f3ee4b4
|
| 3 |
+
size 7157
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.064614037073838e+18,
|
| 4 |
+
"train_loss": 0.8882290616631507,
|
| 5 |
+
"train_runtime": 1944.8842,
|
| 6 |
+
"train_samples_per_second": 84.062,
|
| 7 |
+
"train_steps_per_second": 0.329
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,298 +1,411 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-mobile-eye-tracking-dataset-v2/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"
|
| 14 |
-
"
|
| 15 |
-
"
|
| 16 |
-
"
|
| 17 |
-
|
| 18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
},
|
| 20 |
{
|
| 21 |
-
"epoch":
|
| 22 |
-
"
|
| 23 |
-
"
|
| 24 |
-
"
|
| 25 |
-
"
|
| 26 |
-
"eval_steps_per_second": 0.168,
|
| 27 |
-
"step": 4
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 2.
|
| 31 |
-
"
|
| 32 |
-
"
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
|
| 36 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 37 |
},
|
| 38 |
{
|
| 39 |
"epoch": 4.0,
|
| 40 |
-
"eval_accuracy": 0.
|
| 41 |
-
"eval_loss":
|
| 42 |
-
"eval_runtime":
|
| 43 |
-
"eval_samples_per_second":
|
| 44 |
-
"eval_steps_per_second":
|
| 45 |
-
"step":
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 4.888888888888889,
|
| 49 |
-
"eval_accuracy": 0.5483870967741935,
|
| 50 |
-
"eval_loss": 1.099271535873413,
|
| 51 |
-
"eval_runtime": 5.4311,
|
| 52 |
-
"eval_samples_per_second": 5.708,
|
| 53 |
-
"eval_steps_per_second": 0.184,
|
| 54 |
-
"step": 11
|
| 55 |
-
},
|
| 56 |
-
{
|
| 57 |
-
"epoch": 5.777777777777778,
|
| 58 |
-
"eval_accuracy": 0.6451612903225806,
|
| 59 |
-
"eval_loss": 0.9241538047790527,
|
| 60 |
-
"eval_runtime": 5.4175,
|
| 61 |
-
"eval_samples_per_second": 5.722,
|
| 62 |
-
"eval_steps_per_second": 0.185,
|
| 63 |
-
"step": 13
|
| 64 |
-
},
|
| 65 |
-
{
|
| 66 |
-
"epoch": 6.666666666666667,
|
| 67 |
-
"grad_norm": 4.35577917098999,
|
| 68 |
-
"learning_rate": 4.166666666666667e-05,
|
| 69 |
-
"loss": 1.4667,
|
| 70 |
-
"step": 15
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"epoch":
|
| 74 |
-
"
|
| 75 |
-
"
|
| 76 |
-
"
|
| 77 |
-
"
|
| 78 |
-
|
| 79 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 8.0,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss": 0.
|
| 85 |
-
"eval_runtime": 5.
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second":
|
| 88 |
-
"step":
|
| 89 |
-
},
|
| 90 |
-
{
|
| 91 |
-
"epoch": 8.88888888888889,
|
| 92 |
-
"eval_accuracy": 0.7096774193548387,
|
| 93 |
-
"eval_loss": 0.5325695872306824,
|
| 94 |
-
"eval_runtime": 5.4264,
|
| 95 |
-
"eval_samples_per_second": 5.713,
|
| 96 |
-
"eval_steps_per_second": 0.184,
|
| 97 |
-
"step": 20
|
| 98 |
-
},
|
| 99 |
-
{
|
| 100 |
-
"epoch": 9.777777777777779,
|
| 101 |
-
"eval_accuracy": 0.7419354838709677,
|
| 102 |
-
"eval_loss": 0.48480212688446045,
|
| 103 |
-
"eval_runtime": 5.432,
|
| 104 |
-
"eval_samples_per_second": 5.707,
|
| 105 |
-
"eval_steps_per_second": 0.184,
|
| 106 |
-
"step": 22
|
| 107 |
-
},
|
| 108 |
-
{
|
| 109 |
-
"epoch": 10.666666666666666,
|
| 110 |
-
"eval_accuracy": 0.7741935483870968,
|
| 111 |
-
"eval_loss": 0.4832201302051544,
|
| 112 |
-
"eval_runtime": 5.482,
|
| 113 |
-
"eval_samples_per_second": 5.655,
|
| 114 |
-
"eval_steps_per_second": 0.182,
|
| 115 |
-
"step": 24
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 12.0,
|
| 119 |
-
"eval_accuracy": 0.7741935483870968,
|
| 120 |
-
"eval_loss": 0.44829437136650085,
|
| 121 |
-
"eval_runtime": 5.3703,
|
| 122 |
-
"eval_samples_per_second": 5.772,
|
| 123 |
-
"eval_steps_per_second": 0.186,
|
| 124 |
-
"step": 27
|
| 125 |
-
},
|
| 126 |
-
{
|
| 127 |
-
"epoch": 12.88888888888889,
|
| 128 |
-
"eval_accuracy": 0.7741935483870968,
|
| 129 |
-
"eval_loss": 0.4296128451824188,
|
| 130 |
-
"eval_runtime": 5.4565,
|
| 131 |
-
"eval_samples_per_second": 5.681,
|
| 132 |
-
"eval_steps_per_second": 0.183,
|
| 133 |
-
"step": 29
|
| 134 |
-
},
|
| 135 |
-
{
|
| 136 |
-
"epoch": 13.333333333333334,
|
| 137 |
-
"grad_norm": 5.256907939910889,
|
| 138 |
-
"learning_rate": 2.777777777777778e-05,
|
| 139 |
-
"loss": 0.5925,
|
| 140 |
-
"step": 30
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"epoch":
|
| 144 |
-
"
|
| 145 |
-
"
|
| 146 |
-
"
|
| 147 |
-
"
|
| 148 |
-
"eval_steps_per_second": 0.185,
|
| 149 |
-
"step": 31
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 14.666666666666666,
|
| 153 |
-
"eval_accuracy": 0.8387096774193549,
|
| 154 |
-
"eval_loss": 0.4110867977142334,
|
| 155 |
-
"eval_runtime": 5.4946,
|
| 156 |
-
"eval_samples_per_second": 5.642,
|
| 157 |
-
"eval_steps_per_second": 0.182,
|
| 158 |
-
"step": 33
|
| 159 |
-
},
|
| 160 |
-
{
|
| 161 |
-
"epoch": 16.0,
|
| 162 |
-
"eval_accuracy": 0.8064516129032258,
|
| 163 |
-
"eval_loss": 0.387315571308136,
|
| 164 |
-
"eval_runtime": 5.4222,
|
| 165 |
-
"eval_samples_per_second": 5.717,
|
| 166 |
-
"eval_steps_per_second": 0.184,
|
| 167 |
-
"step": 36
|
| 168 |
-
},
|
| 169 |
-
{
|
| 170 |
-
"epoch": 16.88888888888889,
|
| 171 |
-
"eval_accuracy": 0.8064516129032258,
|
| 172 |
-
"eval_loss": 0.4028545618057251,
|
| 173 |
-
"eval_runtime": 5.4659,
|
| 174 |
-
"eval_samples_per_second": 5.672,
|
| 175 |
-
"eval_steps_per_second": 0.183,
|
| 176 |
-
"step": 38
|
| 177 |
-
},
|
| 178 |
-
{
|
| 179 |
-
"epoch": 17.77777777777778,
|
| 180 |
-
"eval_accuracy": 0.8064516129032258,
|
| 181 |
-
"eval_loss": 0.4065493047237396,
|
| 182 |
-
"eval_runtime": 5.394,
|
| 183 |
-
"eval_samples_per_second": 5.747,
|
| 184 |
-
"eval_steps_per_second": 0.185,
|
| 185 |
-
"step": 40
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 18.666666666666668,
|
| 189 |
-
"eval_accuracy": 0.8064516129032258,
|
| 190 |
-
"eval_loss": 0.38641268014907837,
|
| 191 |
-
"eval_runtime": 5.4499,
|
| 192 |
-
"eval_samples_per_second": 5.688,
|
| 193 |
-
"eval_steps_per_second": 0.183,
|
| 194 |
-
"step": 42
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 20.0,
|
| 198 |
-
"grad_norm": 5.7712812423706055,
|
| 199 |
-
"learning_rate": 1.388888888888889e-05,
|
| 200 |
-
"loss": 0.3285,
|
| 201 |
-
"step": 45
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"
|
| 206 |
-
"
|
| 207 |
-
"
|
| 208 |
-
"
|
| 209 |
-
"eval_steps_per_second": 0.185,
|
| 210 |
-
"step": 45
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"epoch":
|
| 214 |
-
"
|
| 215 |
-
"
|
| 216 |
-
"
|
| 217 |
-
"
|
| 218 |
-
"eval_steps_per_second": 0.187,
|
| 219 |
-
"step": 47
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 21.77777777777778,
|
| 223 |
-
"eval_accuracy": 0.8709677419354839,
|
| 224 |
-
"eval_loss": 0.3871462643146515,
|
| 225 |
-
"eval_runtime": 5.3376,
|
| 226 |
-
"eval_samples_per_second": 5.808,
|
| 227 |
-
"eval_steps_per_second": 0.187,
|
| 228 |
-
"step": 49
|
| 229 |
-
},
|
| 230 |
-
{
|
| 231 |
-
"epoch": 22.666666666666668,
|
| 232 |
-
"eval_accuracy": 0.8064516129032258,
|
| 233 |
-
"eval_loss": 0.3779211640357971,
|
| 234 |
-
"eval_runtime": 5.365,
|
| 235 |
-
"eval_samples_per_second": 5.778,
|
| 236 |
-
"eval_steps_per_second": 0.186,
|
| 237 |
-
"step": 51
|
| 238 |
-
},
|
| 239 |
-
{
|
| 240 |
-
"epoch": 24.0,
|
| 241 |
-
"eval_accuracy": 0.8064516129032258,
|
| 242 |
-
"eval_loss": 0.36982351541519165,
|
| 243 |
-
"eval_runtime": 5.3833,
|
| 244 |
-
"eval_samples_per_second": 5.759,
|
| 245 |
-
"eval_steps_per_second": 0.186,
|
| 246 |
-
"step": 54
|
| 247 |
-
},
|
| 248 |
-
{
|
| 249 |
-
"epoch": 24.88888888888889,
|
| 250 |
-
"eval_accuracy": 0.8387096774193549,
|
| 251 |
-
"eval_loss": 0.37262630462646484,
|
| 252 |
-
"eval_runtime": 5.3782,
|
| 253 |
-
"eval_samples_per_second": 5.764,
|
| 254 |
-
"eval_steps_per_second": 0.186,
|
| 255 |
-
"step": 56
|
| 256 |
-
},
|
| 257 |
-
{
|
| 258 |
-
"epoch": 25.77777777777778,
|
| 259 |
-
"eval_accuracy": 0.8387096774193549,
|
| 260 |
-
"eval_loss": 0.3732232451438904,
|
| 261 |
-
"eval_runtime": 5.4396,
|
| 262 |
-
"eval_samples_per_second": 5.699,
|
| 263 |
-
"eval_steps_per_second": 0.184,
|
| 264 |
-
"step": 58
|
| 265 |
-
},
|
| 266 |
-
{
|
| 267 |
-
"epoch": 26.666666666666668,
|
| 268 |
-
"grad_norm": 3.0856940746307373,
|
| 269 |
-
"learning_rate": 0.0,
|
| 270 |
-
"loss": 0.2621,
|
| 271 |
-
"step": 60
|
| 272 |
},
|
| 273 |
{
|
| 274 |
-
"epoch":
|
| 275 |
-
"
|
| 276 |
-
"
|
| 277 |
-
"
|
| 278 |
-
"
|
| 279 |
-
|
| 280 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"epoch":
|
| 284 |
-
"step":
|
| 285 |
-
"total_flos":
|
| 286 |
-
"train_loss": 0.
|
| 287 |
-
"train_runtime":
|
| 288 |
-
"train_samples_per_second":
|
| 289 |
-
"train_steps_per_second": 0.
|
| 290 |
}
|
| 291 |
],
|
| 292 |
"logging_steps": 15,
|
| 293 |
-
"max_steps":
|
| 294 |
"num_input_tokens_seen": 0,
|
| 295 |
-
"num_train_epochs":
|
| 296 |
"save_steps": 500,
|
| 297 |
"stateful_callbacks": {
|
| 298 |
"TrainerControl": {
|
|
@@ -306,7 +419,7 @@
|
|
| 306 |
"attributes": {}
|
| 307 |
}
|
| 308 |
},
|
| 309 |
-
"total_flos":
|
| 310 |
"train_batch_size": 64,
|
| 311 |
"trial_name": null,
|
| 312 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8257839721254355,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-mobile-eye-tracking-dataset-v2/checkpoint-640",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 640,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.234375,
|
| 13 |
+
"grad_norm": 5.875625133514404,
|
| 14 |
+
"learning_rate": 2.3437500000000002e-06,
|
| 15 |
+
"loss": 2.3858,
|
| 16 |
+
"step": 15
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.46875,
|
| 20 |
+
"grad_norm": 4.278784275054932,
|
| 21 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 22 |
+
"loss": 2.2837,
|
| 23 |
+
"step": 30
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.703125,
|
| 27 |
+
"grad_norm": 4.786104679107666,
|
| 28 |
+
"learning_rate": 7.031250000000001e-06,
|
| 29 |
+
"loss": 2.1191,
|
| 30 |
+
"step": 45
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9375,
|
| 34 |
+
"grad_norm": 4.474501132965088,
|
| 35 |
+
"learning_rate": 9.375000000000001e-06,
|
| 36 |
+
"loss": 1.8826,
|
| 37 |
+
"step": 60
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_accuracy": 0.46689895470383275,
|
| 42 |
+
"eval_loss": 1.5673099756240845,
|
| 43 |
+
"eval_runtime": 9.0893,
|
| 44 |
+
"eval_samples_per_second": 94.727,
|
| 45 |
+
"eval_steps_per_second": 1.54,
|
| 46 |
+
"step": 64
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.171875,
|
| 50 |
+
"grad_norm": 4.6485514640808105,
|
| 51 |
+
"learning_rate": 9.80902777777778e-06,
|
| 52 |
+
"loss": 1.596,
|
| 53 |
+
"step": 75
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.40625,
|
| 57 |
+
"grad_norm": 5.915872573852539,
|
| 58 |
+
"learning_rate": 9.548611111111113e-06,
|
| 59 |
+
"loss": 1.3478,
|
| 60 |
+
"step": 90
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.640625,
|
| 64 |
+
"grad_norm": 11.166051864624023,
|
| 65 |
+
"learning_rate": 9.288194444444444e-06,
|
| 66 |
+
"loss": 1.1995,
|
| 67 |
+
"step": 105
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.875,
|
| 71 |
+
"grad_norm": 8.33702564239502,
|
| 72 |
+
"learning_rate": 9.027777777777779e-06,
|
| 73 |
+
"loss": 1.1123,
|
| 74 |
+
"step": 120
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.0,
|
| 78 |
+
"eval_accuracy": 0.7154471544715447,
|
| 79 |
+
"eval_loss": 0.9031155705451965,
|
| 80 |
+
"eval_runtime": 6.0626,
|
| 81 |
+
"eval_samples_per_second": 142.018,
|
| 82 |
+
"eval_steps_per_second": 2.309,
|
| 83 |
+
"step": 128
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.109375,
|
| 87 |
+
"grad_norm": 9.904069900512695,
|
| 88 |
+
"learning_rate": 8.767361111111112e-06,
|
| 89 |
+
"loss": 1.0379,
|
| 90 |
+
"step": 135
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 2.34375,
|
| 94 |
+
"grad_norm": 6.573493957519531,
|
| 95 |
+
"learning_rate": 8.506944444444445e-06,
|
| 96 |
+
"loss": 0.9758,
|
| 97 |
+
"step": 150
|
|
|
|
|
|
|
| 98 |
},
|
| 99 |
{
|
| 100 |
+
"epoch": 2.578125,
|
| 101 |
+
"grad_norm": 7.493569374084473,
|
| 102 |
+
"learning_rate": 8.246527777777779e-06,
|
| 103 |
+
"loss": 0.9458,
|
| 104 |
+
"step": 165
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.8125,
|
| 108 |
+
"grad_norm": 8.070711135864258,
|
| 109 |
+
"learning_rate": 7.986111111111112e-06,
|
| 110 |
+
"loss": 0.8883,
|
| 111 |
+
"step": 180
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 3.0,
|
| 115 |
+
"eval_accuracy": 0.7572590011614402,
|
| 116 |
+
"eval_loss": 0.7254553437232971,
|
| 117 |
+
"eval_runtime": 5.0,
|
| 118 |
+
"eval_samples_per_second": 172.198,
|
| 119 |
+
"eval_steps_per_second": 2.8,
|
| 120 |
+
"step": 192
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.046875,
|
| 124 |
+
"grad_norm": 6.530824661254883,
|
| 125 |
+
"learning_rate": 7.725694444444445e-06,
|
| 126 |
+
"loss": 0.8797,
|
| 127 |
+
"step": 195
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.28125,
|
| 131 |
+
"grad_norm": 4.865982532501221,
|
| 132 |
+
"learning_rate": 7.465277777777778e-06,
|
| 133 |
+
"loss": 0.8543,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.515625,
|
| 138 |
+
"grad_norm": 8.852520942687988,
|
| 139 |
+
"learning_rate": 7.204861111111112e-06,
|
| 140 |
+
"loss": 0.8026,
|
| 141 |
+
"step": 225
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.75,
|
| 145 |
+
"grad_norm": 10.359617233276367,
|
| 146 |
+
"learning_rate": 6.944444444444445e-06,
|
| 147 |
+
"loss": 0.7792,
|
| 148 |
+
"step": 240
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.984375,
|
| 152 |
+
"grad_norm": 8.156173706054688,
|
| 153 |
+
"learning_rate": 6.684027777777779e-06,
|
| 154 |
+
"loss": 0.7778,
|
| 155 |
+
"step": 255
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 4.0,
|
| 159 |
+
"eval_accuracy": 0.7793263646922184,
|
| 160 |
+
"eval_loss": 0.621900737285614,
|
| 161 |
+
"eval_runtime": 4.9966,
|
| 162 |
+
"eval_samples_per_second": 172.316,
|
| 163 |
+
"eval_steps_per_second": 2.802,
|
| 164 |
+
"step": 256
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 4.21875,
|
| 168 |
+
"grad_norm": 7.673396587371826,
|
| 169 |
+
"learning_rate": 6.423611111111112e-06,
|
| 170 |
+
"loss": 0.7533,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.453125,
|
| 175 |
+
"grad_norm": 7.9218645095825195,
|
| 176 |
+
"learning_rate": 6.163194444444444e-06,
|
| 177 |
+
"loss": 0.7165,
|
| 178 |
+
"step": 285
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.6875,
|
| 182 |
+
"grad_norm": 5.629858493804932,
|
| 183 |
+
"learning_rate": 5.9027777777777785e-06,
|
| 184 |
+
"loss": 0.7092,
|
| 185 |
+
"step": 300
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.921875,
|
| 189 |
+
"grad_norm": 9.611222267150879,
|
| 190 |
+
"learning_rate": 5.642361111111112e-06,
|
| 191 |
+
"loss": 0.708,
|
| 192 |
+
"step": 315
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 5.0,
|
| 196 |
+
"eval_accuracy": 0.8002322880371661,
|
| 197 |
+
"eval_loss": 0.5520657896995544,
|
| 198 |
+
"eval_runtime": 4.9527,
|
| 199 |
+
"eval_samples_per_second": 173.845,
|
| 200 |
+
"eval_steps_per_second": 2.827,
|
| 201 |
+
"step": 320
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.15625,
|
| 205 |
+
"grad_norm": 8.07551383972168,
|
| 206 |
+
"learning_rate": 5.381944444444445e-06,
|
| 207 |
+
"loss": 0.6842,
|
| 208 |
+
"step": 330
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.390625,
|
| 212 |
+
"grad_norm": 7.263919830322266,
|
| 213 |
+
"learning_rate": 5.121527777777778e-06,
|
| 214 |
+
"loss": 0.6724,
|
| 215 |
+
"step": 345
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.625,
|
| 219 |
+
"grad_norm": 6.807614326477051,
|
| 220 |
+
"learning_rate": 4.861111111111111e-06,
|
| 221 |
+
"loss": 0.6703,
|
| 222 |
+
"step": 360
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.859375,
|
| 226 |
+
"grad_norm": 6.808135509490967,
|
| 227 |
+
"learning_rate": 4.6006944444444446e-06,
|
| 228 |
+
"loss": 0.6308,
|
| 229 |
+
"step": 375
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 6.0,
|
| 233 |
+
"eval_accuracy": 0.8130081300813008,
|
| 234 |
+
"eval_loss": 0.5193033218383789,
|
| 235 |
+
"eval_runtime": 5.1321,
|
| 236 |
+
"eval_samples_per_second": 167.766,
|
| 237 |
+
"eval_steps_per_second": 2.728,
|
| 238 |
+
"step": 384
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.09375,
|
| 242 |
+
"grad_norm": 6.225130081176758,
|
| 243 |
+
"learning_rate": 4.340277777777779e-06,
|
| 244 |
+
"loss": 0.6442,
|
| 245 |
+
"step": 390
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.328125,
|
| 249 |
+
"grad_norm": 9.285326957702637,
|
| 250 |
+
"learning_rate": 4.079861111111111e-06,
|
| 251 |
+
"loss": 0.6139,
|
| 252 |
+
"step": 405
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 6.5625,
|
| 256 |
+
"grad_norm": 7.4434356689453125,
|
| 257 |
+
"learning_rate": 3.819444444444444e-06,
|
| 258 |
+
"loss": 0.6164,
|
| 259 |
+
"step": 420
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 6.796875,
|
| 263 |
+
"grad_norm": 7.437946796417236,
|
| 264 |
+
"learning_rate": 3.5590277777777783e-06,
|
| 265 |
+
"loss": 0.6142,
|
| 266 |
+
"step": 435
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 7.0,
|
| 270 |
+
"eval_accuracy": 0.8234610917537747,
|
| 271 |
+
"eval_loss": 0.4853927195072174,
|
| 272 |
+
"eval_runtime": 5.3739,
|
| 273 |
+
"eval_samples_per_second": 160.219,
|
| 274 |
+
"eval_steps_per_second": 2.605,
|
| 275 |
+
"step": 448
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.03125,
|
| 279 |
+
"grad_norm": 8.036016464233398,
|
| 280 |
+
"learning_rate": 3.2986111111111115e-06,
|
| 281 |
+
"loss": 0.6178,
|
| 282 |
+
"step": 450
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.265625,
|
| 286 |
+
"grad_norm": 6.326539039611816,
|
| 287 |
+
"learning_rate": 3.0381944444444443e-06,
|
| 288 |
+
"loss": 0.6076,
|
| 289 |
+
"step": 465
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.5,
|
| 293 |
+
"grad_norm": 6.9872236251831055,
|
| 294 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 295 |
+
"loss": 0.5688,
|
| 296 |
+
"step": 480
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.734375,
|
| 300 |
+
"grad_norm": 7.621065139770508,
|
| 301 |
+
"learning_rate": 2.517361111111111e-06,
|
| 302 |
+
"loss": 0.6034,
|
| 303 |
+
"step": 495
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 7.96875,
|
| 307 |
+
"grad_norm": 7.739367485046387,
|
| 308 |
+
"learning_rate": 2.2569444444444448e-06,
|
| 309 |
+
"loss": 0.5817,
|
| 310 |
+
"step": 510
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 8.0,
|
| 314 |
+
"eval_accuracy": 0.8199767711962834,
|
| 315 |
+
"eval_loss": 0.472567081451416,
|
| 316 |
+
"eval_runtime": 5.1622,
|
| 317 |
+
"eval_samples_per_second": 166.788,
|
| 318 |
+
"eval_steps_per_second": 2.712,
|
| 319 |
+
"step": 512
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 8.203125,
|
| 323 |
+
"grad_norm": 7.276366710662842,
|
| 324 |
+
"learning_rate": 1.996527777777778e-06,
|
| 325 |
+
"loss": 0.5682,
|
| 326 |
+
"step": 525
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 8.4375,
|
| 330 |
+
"grad_norm": 9.095114707946777,
|
| 331 |
+
"learning_rate": 1.7361111111111112e-06,
|
| 332 |
+
"loss": 0.5797,
|
| 333 |
+
"step": 540
|
|
|
|
|
|
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 8.671875,
|
| 337 |
+
"grad_norm": 7.369579792022705,
|
| 338 |
+
"learning_rate": 1.4756944444444446e-06,
|
| 339 |
+
"loss": 0.5732,
|
| 340 |
+
"step": 555
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 8.90625,
|
| 344 |
+
"grad_norm": 6.826248645782471,
|
| 345 |
+
"learning_rate": 1.2152777777777778e-06,
|
| 346 |
+
"loss": 0.5952,
|
| 347 |
+
"step": 570
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 9.0,
|
| 351 |
+
"eval_accuracy": 0.8211382113821138,
|
| 352 |
+
"eval_loss": 0.46478894352912903,
|
| 353 |
+
"eval_runtime": 5.0463,
|
| 354 |
+
"eval_samples_per_second": 170.621,
|
| 355 |
+
"eval_steps_per_second": 2.774,
|
| 356 |
+
"step": 576
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.140625,
|
| 360 |
+
"grad_norm": 12.532827377319336,
|
| 361 |
+
"learning_rate": 9.54861111111111e-07,
|
| 362 |
+
"loss": 0.5849,
|
| 363 |
+
"step": 585
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 9.375,
|
| 367 |
+
"grad_norm": 8.84656047821045,
|
| 368 |
+
"learning_rate": 6.944444444444446e-07,
|
| 369 |
+
"loss": 0.583,
|
| 370 |
+
"step": 600
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 9.609375,
|
| 374 |
+
"grad_norm": 9.143095970153809,
|
| 375 |
+
"learning_rate": 4.340277777777778e-07,
|
| 376 |
+
"loss": 0.5599,
|
| 377 |
+
"step": 615
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 9.84375,
|
| 381 |
+
"grad_norm": 7.61570930480957,
|
| 382 |
+
"learning_rate": 1.7361111111111115e-07,
|
| 383 |
+
"loss": 0.5915,
|
| 384 |
+
"step": 630
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 10.0,
|
| 388 |
+
"eval_accuracy": 0.8257839721254355,
|
| 389 |
+
"eval_loss": 0.4613528847694397,
|
| 390 |
+
"eval_runtime": 5.2149,
|
| 391 |
+
"eval_samples_per_second": 165.103,
|
| 392 |
+
"eval_steps_per_second": 2.685,
|
| 393 |
+
"step": 640
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 10.0,
|
| 397 |
+
"step": 640,
|
| 398 |
+
"total_flos": 4.064614037073838e+18,
|
| 399 |
+
"train_loss": 0.8882290616631507,
|
| 400 |
+
"train_runtime": 1944.8842,
|
| 401 |
+
"train_samples_per_second": 84.062,
|
| 402 |
+
"train_steps_per_second": 0.329
|
| 403 |
}
|
| 404 |
],
|
| 405 |
"logging_steps": 15,
|
| 406 |
+
"max_steps": 640,
|
| 407 |
"num_input_tokens_seen": 0,
|
| 408 |
+
"num_train_epochs": 10,
|
| 409 |
"save_steps": 500,
|
| 410 |
"stateful_callbacks": {
|
| 411 |
"TrainerControl": {
|
|
|
|
| 419 |
"attributes": {}
|
| 420 |
}
|
| 421 |
},
|
| 422 |
+
"total_flos": 4.064614037073838e+18,
|
| 423 |
"train_batch_size": 64,
|
| 424 |
"trial_name": null,
|
| 425 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e757de8f4fdb59c52b6ad3219481d0b6bb84b344ae9ca28d91a978cac817f0ea
|
| 3 |
size 5304
|