Commit
·
ebef2ad
1
Parent(s):
45b98b5
Training in progress, step 200
Browse files- all_results.json +16 -16
- eval_results.json +6 -6
- pytorch_model.bin +2 -2
- runs/Feb01_09-54-08_node1/events.out.tfevents.1643680621.node1 +2 -2
- runs/Feb01_09-54-08_node1/events.out.tfevents.1643694057.node1 +3 -0
- runs/Feb02_21-25-57_node1/1643808618.960355/events.out.tfevents.1643808618.node1 +3 -0
- runs/Feb02_21-25-57_node1/events.out.tfevents.1643808618.node1 +3 -0
- test_results.json +6 -6
- tokenizer.json +0 -0
- train_results.json +4 -4
- trainer_state.json +2002 -230
- training_args.bin +1 -1
- training_config.json +1 -1
all_results.json
CHANGED
|
@@ -1,20 +1,20 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"eval_accuracy":
|
| 4 |
-
"eval_average_metrics": 94.
|
| 5 |
-
"eval_f1": 93.
|
| 6 |
-
"eval_loss": 0.
|
| 7 |
-
"eval_runtime":
|
| 8 |
-
"eval_samples_per_second":
|
| 9 |
-
"test_accuracy": 90.
|
| 10 |
-
"test_average_metrics": 88.
|
| 11 |
-
"test_f1":
|
| 12 |
-
"test_loss": 0.
|
| 13 |
-
"test_runtime":
|
| 14 |
-
"test_samples_per_second":
|
| 15 |
-
"train_loss": 0.
|
| 16 |
-
"train_runtime":
|
| 17 |
"train_samples": 362846,
|
| 18 |
-
"train_samples_per_second":
|
| 19 |
-
"train_steps_per_second": 2.
|
| 20 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 95.0,
|
| 4 |
+
"eval_average_metrics": 94.22774869109946,
|
| 5 |
+
"eval_f1": 93.45549738219894,
|
| 6 |
+
"eval_loss": 0.055720701813697815,
|
| 7 |
+
"eval_runtime": 4.796,
|
| 8 |
+
"eval_samples_per_second": 208.506,
|
| 9 |
+
"test_accuracy": 90.3141231758595,
|
| 10 |
+
"test_average_metrics": 88.72727044120705,
|
| 11 |
+
"test_f1": 87.14041770655459,
|
| 12 |
+
"test_loss": 0.09215616434812546,
|
| 13 |
+
"test_runtime": 180.597,
|
| 14 |
+
"test_samples_per_second": 223.869,
|
| 15 |
+
"train_loss": 0.07169761398949699,
|
| 16 |
+
"train_runtime": 13428.6442,
|
| 17 |
"train_samples": 362846,
|
| 18 |
+
"train_samples_per_second": 81.061,
|
| 19 |
+
"train_steps_per_second": 2.533
|
| 20 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"eval_accuracy":
|
| 4 |
-
"eval_average_metrics": 94.
|
| 5 |
-
"eval_f1": 93.
|
| 6 |
-
"eval_loss": 0.
|
| 7 |
-
"eval_runtime":
|
| 8 |
-
"eval_samples_per_second":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"eval_accuracy": 95.0,
|
| 4 |
+
"eval_average_metrics": 94.22774869109946,
|
| 5 |
+
"eval_f1": 93.45549738219894,
|
| 6 |
+
"eval_loss": 0.055720701813697815,
|
| 7 |
+
"eval_runtime": 4.796,
|
| 8 |
+
"eval_samples_per_second": 208.506
|
| 9 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:408f9f69d04647ca9fc02f4fa700ca2a77bdf482d046051f409a9a93fdd80734
|
| 3 |
+
size 2631685
|
runs/Feb01_09-54-08_node1/events.out.tfevents.1643680621.node1
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f282ed9800579880d127ed781c6242319b0476e61a79e8f0d3b752afba87760a
|
| 3 |
+
size 78623
|
runs/Feb01_09-54-08_node1/events.out.tfevents.1643694057.node1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b3dcfdd25f487c2cedc8f5528fa291645116206e9fe02731aa1aa743afdc1a1
|
| 3 |
+
size 790
|
runs/Feb02_21-25-57_node1/1643808618.960355/events.out.tfevents.1643808618.node1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7db4374f8c0dacd11d450cec2ad2b6516845e7b69d5077dfed97bc745676d76c
|
| 3 |
+
size 5008
|
runs/Feb02_21-25-57_node1/events.out.tfevents.1643808618.node1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83e060981f177e86172998af1f56d4555bac0e739cc33f88dd0a60aa187f6f4e
|
| 3 |
+
size 4296
|
test_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"test_accuracy": 90.
|
| 4 |
-
"test_average_metrics": 88.
|
| 5 |
-
"test_f1":
|
| 6 |
-
"test_loss": 0.
|
| 7 |
-
"test_runtime":
|
| 8 |
-
"test_samples_per_second":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"test_accuracy": 90.3141231758595,
|
| 4 |
+
"test_average_metrics": 88.72727044120705,
|
| 5 |
+
"test_f1": 87.14041770655459,
|
| 6 |
+
"test_loss": 0.09215616434812546,
|
| 7 |
+
"test_runtime": 180.597,
|
| 8 |
+
"test_samples_per_second": 223.869
|
| 9 |
}
|
tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 362846,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 2.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.0,
|
| 3 |
+
"train_loss": 0.07169761398949699,
|
| 4 |
+
"train_runtime": 13428.6442,
|
| 5 |
"train_samples": 362846,
|
| 6 |
+
"train_samples_per_second": 81.061,
|
| 7 |
+
"train_steps_per_second": 2.533
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,361 +1,2133 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 94.
|
| 3 |
-
"best_model_checkpoint": "outputs/bitfit/t5-base/qqp/checkpoint-
|
| 4 |
"epoch": 3.0,
|
| 5 |
-
"global_step":
|
| 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.
|
| 12 |
-
"
|
| 13 |
-
"
|
| 14 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"eval_accuracy":
|
| 19 |
-
"eval_average_metrics":
|
| 20 |
-
"eval_f1":
|
| 21 |
-
"eval_loss": 0.
|
| 22 |
-
"eval_runtime":
|
| 23 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
"step": 500
|
| 25 |
},
|
| 26 |
{
|
| 27 |
-
"epoch": 0.
|
| 28 |
-
"
|
| 29 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 1000
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"eval_accuracy":
|
| 35 |
-
"eval_average_metrics":
|
| 36 |
-
"eval_f1":
|
| 37 |
-
"eval_loss": 0.
|
| 38 |
-
"eval_runtime":
|
| 39 |
-
"eval_samples_per_second":
|
| 40 |
"step": 1000
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"epoch": 0.
|
| 44 |
-
"
|
| 45 |
-
"
|
| 46 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"epoch": 0.
|
| 50 |
-
"eval_accuracy": 93.
|
| 51 |
-
"eval_average_metrics": 92.
|
| 52 |
-
"eval_f1":
|
| 53 |
-
"eval_loss": 0.
|
| 54 |
-
"eval_runtime":
|
| 55 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
"step": 1500
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
-
"
|
| 61 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
"step": 2000
|
| 63 |
},
|
| 64 |
{
|
| 65 |
-
"epoch": 0.
|
| 66 |
-
"eval_accuracy":
|
| 67 |
-
"eval_average_metrics":
|
| 68 |
-
"eval_f1":
|
| 69 |
-
"eval_loss": 0.
|
| 70 |
-
"eval_runtime":
|
| 71 |
-
"eval_samples_per_second":
|
| 72 |
"step": 2000
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"
|
| 77 |
-
"
|
| 78 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch": 0.
|
| 82 |
-
"eval_accuracy":
|
| 83 |
-
"eval_average_metrics":
|
| 84 |
-
"eval_f1":
|
| 85 |
-
"eval_loss": 0.
|
| 86 |
-
"eval_runtime":
|
| 87 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
"step": 2500
|
| 89 |
},
|
| 90 |
{
|
| 91 |
-
"epoch": 0.
|
| 92 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
"loss": 0.0764,
|
| 94 |
"step": 3000
|
| 95 |
},
|
| 96 |
{
|
| 97 |
-
"epoch": 0.
|
| 98 |
-
"eval_accuracy": 94.
|
| 99 |
-
"eval_average_metrics": 93.
|
| 100 |
-
"eval_f1": 92.
|
| 101 |
-
"eval_loss": 0.
|
| 102 |
-
"eval_runtime":
|
| 103 |
-
"eval_samples_per_second":
|
| 104 |
"step": 3000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 0.
|
| 108 |
-
"
|
| 109 |
-
"
|
| 110 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 0.
|
| 114 |
-
"eval_accuracy": 93.
|
| 115 |
-
"eval_average_metrics":
|
| 116 |
-
"eval_f1":
|
| 117 |
-
"eval_loss": 0.
|
| 118 |
-
"eval_runtime":
|
| 119 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 120 |
"step": 3500
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch":
|
| 124 |
-
"
|
| 125 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 126 |
"step": 4000
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch":
|
| 130 |
-
"eval_accuracy":
|
| 131 |
-
"eval_average_metrics":
|
| 132 |
-
"eval_f1":
|
| 133 |
-
"eval_loss": 0.
|
| 134 |
-
"eval_runtime":
|
| 135 |
-
"eval_samples_per_second":
|
| 136 |
"step": 4000
|
| 137 |
},
|
| 138 |
{
|
| 139 |
-
"epoch":
|
| 140 |
-
"
|
| 141 |
-
"
|
| 142 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch":
|
| 146 |
-
"eval_accuracy":
|
| 147 |
-
"eval_average_metrics":
|
| 148 |
-
"eval_f1":
|
| 149 |
-
"eval_loss": 0.
|
| 150 |
-
"eval_runtime":
|
| 151 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 152 |
"step": 4500
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch":
|
| 156 |
-
"
|
| 157 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
"step": 5000
|
| 159 |
},
|
| 160 |
{
|
| 161 |
-
"epoch":
|
| 162 |
-
"eval_accuracy":
|
| 163 |
-
"eval_average_metrics":
|
| 164 |
-
"eval_f1":
|
| 165 |
-
"eval_loss": 0.
|
| 166 |
-
"eval_runtime":
|
| 167 |
-
"eval_samples_per_second":
|
| 168 |
"step": 5000
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch":
|
| 172 |
-
"
|
| 173 |
-
"
|
| 174 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch":
|
| 178 |
-
"eval_accuracy":
|
| 179 |
-
"eval_average_metrics":
|
| 180 |
-
"eval_f1":
|
| 181 |
-
"eval_loss": 0.
|
| 182 |
-
"eval_runtime":
|
| 183 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 184 |
"step": 5500
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch":
|
| 188 |
-
"
|
| 189 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 190 |
"step": 6000
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"epoch":
|
| 194 |
-
"eval_accuracy": 94.
|
| 195 |
-
"eval_average_metrics": 93.
|
| 196 |
-
"eval_f1": 92.
|
| 197 |
-
"eval_loss": 0.
|
| 198 |
-
"eval_runtime":
|
| 199 |
-
"eval_samples_per_second":
|
| 200 |
"step": 6000
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch":
|
| 204 |
-
"
|
| 205 |
-
"
|
| 206 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
-
"eval_accuracy":
|
| 211 |
-
"eval_average_metrics":
|
| 212 |
-
"eval_f1":
|
| 213 |
-
"eval_loss": 0.
|
| 214 |
-
"eval_runtime":
|
| 215 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 216 |
"step": 6500
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch":
|
| 220 |
-
"
|
| 221 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 222 |
"step": 7000
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch":
|
| 226 |
-
"eval_accuracy": 94.
|
| 227 |
-
"eval_average_metrics": 93.
|
| 228 |
-
"eval_f1": 92.
|
| 229 |
-
"eval_loss": 0.
|
| 230 |
-
"eval_runtime":
|
| 231 |
-
"eval_samples_per_second":
|
| 232 |
"step": 7000
|
| 233 |
},
|
| 234 |
{
|
| 235 |
-
"epoch":
|
| 236 |
-
"
|
| 237 |
-
"
|
| 238 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 239 |
},
|
| 240 |
{
|
| 241 |
-
"epoch":
|
| 242 |
-
"eval_accuracy": 94.
|
| 243 |
-
"eval_average_metrics": 93.
|
| 244 |
-
"eval_f1": 92.
|
| 245 |
-
"eval_loss": 0.
|
| 246 |
-
"eval_runtime":
|
| 247 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 248 |
"step": 7500
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"
|
| 253 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 254 |
"step": 8000
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"eval_accuracy": 94.
|
| 259 |
-
"eval_average_metrics": 93.
|
| 260 |
-
"eval_f1": 92.
|
| 261 |
-
"eval_loss": 0.
|
| 262 |
-
"eval_runtime":
|
| 263 |
-
"eval_samples_per_second":
|
| 264 |
"step": 8000
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch":
|
| 268 |
-
"
|
| 269 |
-
"
|
| 270 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"eval_accuracy": 94.
|
| 275 |
-
"eval_average_metrics": 93.
|
| 276 |
-
"eval_f1": 92.
|
| 277 |
-
"eval_loss": 0.
|
| 278 |
-
"eval_runtime":
|
| 279 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 280 |
"step": 8500
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"epoch":
|
| 284 |
-
"
|
| 285 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 286 |
"step": 9000
|
| 287 |
},
|
| 288 |
{
|
| 289 |
-
"epoch":
|
| 290 |
-
"eval_accuracy":
|
| 291 |
-
"eval_average_metrics":
|
| 292 |
-
"eval_f1":
|
| 293 |
-
"eval_loss": 0.
|
| 294 |
-
"eval_runtime":
|
| 295 |
-
"eval_samples_per_second":
|
| 296 |
"step": 9000
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch":
|
| 300 |
-
"
|
| 301 |
-
"
|
| 302 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch":
|
| 306 |
"eval_accuracy": 94.39999999999999,
|
| 307 |
-
"eval_average_metrics": 93.
|
| 308 |
-
"eval_f1": 92.
|
| 309 |
-
"eval_loss": 0.
|
| 310 |
-
"eval_runtime":
|
| 311 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 312 |
"step": 9500
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"epoch":
|
| 316 |
-
"
|
| 317 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 318 |
"step": 10000
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch":
|
| 322 |
-
"eval_accuracy":
|
| 323 |
-
"eval_average_metrics":
|
| 324 |
-
"eval_f1": 92.
|
| 325 |
-
"eval_loss": 0.
|
| 326 |
-
"eval_runtime":
|
| 327 |
-
"eval_samples_per_second":
|
| 328 |
"step": 10000
|
| 329 |
},
|
| 330 |
{
|
| 331 |
-
"epoch":
|
| 332 |
-
"
|
| 333 |
-
"
|
| 334 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 335 |
},
|
| 336 |
{
|
| 337 |
-
"epoch":
|
| 338 |
-
"eval_accuracy": 94.
|
| 339 |
-
"eval_average_metrics": 93.
|
| 340 |
-
"eval_f1":
|
| 341 |
-
"eval_loss": 0.
|
| 342 |
-
"eval_runtime":
|
| 343 |
-
"eval_samples_per_second":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 344 |
"step": 10500
|
| 345 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 346 |
{
|
| 347 |
"epoch": 3.0,
|
| 348 |
-
"step":
|
| 349 |
-
"total_flos": 1.
|
| 350 |
-
"train_loss": 0.
|
| 351 |
-
"train_runtime":
|
| 352 |
-
"train_samples_per_second":
|
| 353 |
-
"train_steps_per_second": 2.
|
| 354 |
}
|
| 355 |
],
|
| 356 |
-
"max_steps":
|
| 357 |
"num_train_epochs": 3,
|
| 358 |
-
"total_flos": 1.
|
| 359 |
"trial_name": null,
|
| 360 |
"trial_params": null
|
| 361 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 94.22774869109946,
|
| 3 |
+
"best_model_checkpoint": "outputs/bitfit/t5-base/qqp/checkpoint-8600",
|
| 4 |
"epoch": 3.0,
|
| 5 |
+
"global_step": 34017,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"eval_accuracy": 93.60000000000001,
|
| 13 |
+
"eval_average_metrics": 92.41643835616439,
|
| 14 |
+
"eval_f1": 91.23287671232877,
|
| 15 |
+
"eval_loss": 0.06451932340860367,
|
| 16 |
+
"eval_runtime": 4.5218,
|
| 17 |
+
"eval_samples_per_second": 221.15,
|
| 18 |
+
"step": 200
|
| 19 |
},
|
| 20 |
{
|
| 21 |
+
"epoch": 0.04,
|
| 22 |
+
"eval_accuracy": 90.2,
|
| 23 |
+
"eval_average_metrics": 89.15339805825244,
|
| 24 |
+
"eval_f1": 88.10679611650487,
|
| 25 |
+
"eval_loss": 0.08826350420713425,
|
| 26 |
+
"eval_runtime": 4.5364,
|
| 27 |
+
"eval_samples_per_second": 220.438,
|
| 28 |
+
"step": 400
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.04,
|
| 32 |
+
"learning_rate": 0.00029559044007408056,
|
| 33 |
+
"loss": 0.1448,
|
| 34 |
"step": 500
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"epoch": 0.05,
|
| 38 |
+
"eval_accuracy": 93.7,
|
| 39 |
+
"eval_average_metrics": 92.76439688715953,
|
| 40 |
+
"eval_f1": 91.82879377431907,
|
| 41 |
+
"eval_loss": 0.061929114162921906,
|
| 42 |
+
"eval_runtime": 4.5766,
|
| 43 |
+
"eval_samples_per_second": 218.503,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07,
|
| 48 |
+
"eval_accuracy": 93.30000000000001,
|
| 49 |
+
"eval_average_metrics": 92.3385456885457,
|
| 50 |
+
"eval_f1": 91.37709137709138,
|
| 51 |
+
"eval_loss": 0.0669504851102829,
|
| 52 |
+
"eval_runtime": 4.5237,
|
| 53 |
+
"eval_samples_per_second": 221.059,
|
| 54 |
+
"step": 800
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09,
|
| 58 |
+
"learning_rate": 0.0002911808801481612,
|
| 59 |
+
"loss": 0.0782,
|
| 60 |
"step": 1000
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"epoch": 0.09,
|
| 64 |
+
"eval_accuracy": 93.7,
|
| 65 |
+
"eval_average_metrics": 92.74308996088658,
|
| 66 |
+
"eval_f1": 91.78617992177314,
|
| 67 |
+
"eval_loss": 0.06603064388036728,
|
| 68 |
+
"eval_runtime": 4.727,
|
| 69 |
+
"eval_samples_per_second": 211.549,
|
| 70 |
"step": 1000
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"epoch": 0.11,
|
| 74 |
+
"eval_accuracy": 93.30000000000001,
|
| 75 |
+
"eval_average_metrics": 92.14125168236878,
|
| 76 |
+
"eval_f1": 90.98250336473754,
|
| 77 |
+
"eval_loss": 0.06394415348768234,
|
| 78 |
+
"eval_runtime": 4.6179,
|
| 79 |
+
"eval_samples_per_second": 216.548,
|
| 80 |
+
"step": 1200
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 0.12,
|
| 84 |
+
"eval_accuracy": 93.30000000000001,
|
| 85 |
+
"eval_average_metrics": 92.07974079126876,
|
| 86 |
+
"eval_f1": 90.85948158253751,
|
| 87 |
+
"eval_loss": 0.059915054589509964,
|
| 88 |
+
"eval_runtime": 4.4912,
|
| 89 |
+
"eval_samples_per_second": 222.655,
|
| 90 |
+
"step": 1400
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.13,
|
| 94 |
+
"learning_rate": 0.0002867713202222418,
|
| 95 |
+
"loss": 0.0773,
|
| 96 |
"step": 1500
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 0.14,
|
| 100 |
+
"eval_accuracy": 93.0,
|
| 101 |
+
"eval_average_metrics": 92.0242966751918,
|
| 102 |
+
"eval_f1": 91.04859335038363,
|
| 103 |
+
"eval_loss": 0.06896140426397324,
|
| 104 |
+
"eval_runtime": 4.614,
|
| 105 |
+
"eval_samples_per_second": 216.732,
|
| 106 |
+
"step": 1600
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.16,
|
| 110 |
+
"eval_accuracy": 93.60000000000001,
|
| 111 |
+
"eval_average_metrics": 92.61151832460735,
|
| 112 |
+
"eval_f1": 91.62303664921467,
|
| 113 |
+
"eval_loss": 0.06234096363186836,
|
| 114 |
+
"eval_runtime": 4.6818,
|
| 115 |
+
"eval_samples_per_second": 213.593,
|
| 116 |
+
"step": 1800
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.18,
|
| 120 |
+
"learning_rate": 0.0002823617602963224,
|
| 121 |
+
"loss": 0.0746,
|
| 122 |
"step": 2000
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.18,
|
| 126 |
+
"eval_accuracy": 94.19999999999999,
|
| 127 |
+
"eval_average_metrics": 93.23333333333332,
|
| 128 |
+
"eval_f1": 92.26666666666667,
|
| 129 |
+
"eval_loss": 0.0571160614490509,
|
| 130 |
+
"eval_runtime": 4.7897,
|
| 131 |
+
"eval_samples_per_second": 208.783,
|
| 132 |
"step": 2000
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 0.19,
|
| 136 |
+
"eval_accuracy": 93.60000000000001,
|
| 137 |
+
"eval_average_metrics": 92.54468085106384,
|
| 138 |
+
"eval_f1": 91.48936170212767,
|
| 139 |
+
"eval_loss": 0.059621669352054596,
|
| 140 |
+
"eval_runtime": 4.6087,
|
| 141 |
+
"eval_samples_per_second": 216.979,
|
| 142 |
+
"step": 2200
|
| 143 |
},
|
| 144 |
{
|
| 145 |
+
"epoch": 0.21,
|
| 146 |
+
"eval_accuracy": 93.60000000000001,
|
| 147 |
+
"eval_average_metrics": 92.36786703601109,
|
| 148 |
+
"eval_f1": 91.13573407202216,
|
| 149 |
+
"eval_loss": 0.06162749230861664,
|
| 150 |
+
"eval_runtime": 4.7312,
|
| 151 |
+
"eval_samples_per_second": 211.361,
|
| 152 |
+
"step": 2400
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.22,
|
| 156 |
+
"learning_rate": 0.00027795220037040303,
|
| 157 |
+
"loss": 0.0771,
|
| 158 |
"step": 2500
|
| 159 |
},
|
| 160 |
{
|
| 161 |
+
"epoch": 0.23,
|
| 162 |
+
"eval_accuracy": 93.10000000000001,
|
| 163 |
+
"eval_average_metrics": 91.91912751677853,
|
| 164 |
+
"eval_f1": 90.73825503355705,
|
| 165 |
+
"eval_loss": 0.06196223199367523,
|
| 166 |
+
"eval_runtime": 4.4778,
|
| 167 |
+
"eval_samples_per_second": 223.322,
|
| 168 |
+
"step": 2600
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.25,
|
| 172 |
+
"eval_accuracy": 93.7,
|
| 173 |
+
"eval_average_metrics": 92.72155963302754,
|
| 174 |
+
"eval_f1": 91.74311926605506,
|
| 175 |
+
"eval_loss": 0.059104129672050476,
|
| 176 |
+
"eval_runtime": 4.462,
|
| 177 |
+
"eval_samples_per_second": 224.113,
|
| 178 |
+
"step": 2800
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.26,
|
| 182 |
+
"learning_rate": 0.0002735426404444836,
|
| 183 |
"loss": 0.0764,
|
| 184 |
"step": 3000
|
| 185 |
},
|
| 186 |
{
|
| 187 |
+
"epoch": 0.26,
|
| 188 |
+
"eval_accuracy": 94.1,
|
| 189 |
+
"eval_average_metrics": 93.1633069828722,
|
| 190 |
+
"eval_f1": 92.2266139657444,
|
| 191 |
+
"eval_loss": 0.059123676270246506,
|
| 192 |
+
"eval_runtime": 4.5787,
|
| 193 |
+
"eval_samples_per_second": 218.4,
|
| 194 |
"step": 3000
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 0.28,
|
| 198 |
+
"eval_accuracy": 91.5,
|
| 199 |
+
"eval_average_metrics": 90.49660074165637,
|
| 200 |
+
"eval_f1": 89.49320148331273,
|
| 201 |
+
"eval_loss": 0.07883985340595245,
|
| 202 |
+
"eval_runtime": 4.5148,
|
| 203 |
+
"eval_samples_per_second": 221.496,
|
| 204 |
+
"step": 3200
|
| 205 |
},
|
| 206 |
{
|
| 207 |
+
"epoch": 0.3,
|
| 208 |
+
"eval_accuracy": 93.10000000000001,
|
| 209 |
+
"eval_average_metrics": 92.02837483617301,
|
| 210 |
+
"eval_f1": 90.956749672346,
|
| 211 |
+
"eval_loss": 0.06356123834848404,
|
| 212 |
+
"eval_runtime": 4.5836,
|
| 213 |
+
"eval_samples_per_second": 218.171,
|
| 214 |
+
"step": 3400
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.31,
|
| 218 |
+
"learning_rate": 0.0002691330805185642,
|
| 219 |
+
"loss": 0.0732,
|
| 220 |
"step": 3500
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"epoch": 0.32,
|
| 224 |
+
"eval_accuracy": 93.7,
|
| 225 |
+
"eval_average_metrics": 92.598987854251,
|
| 226 |
+
"eval_f1": 91.49797570850201,
|
| 227 |
+
"eval_loss": 0.058661118149757385,
|
| 228 |
+
"eval_runtime": 4.581,
|
| 229 |
+
"eval_samples_per_second": 218.292,
|
| 230 |
+
"step": 3600
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.34,
|
| 234 |
+
"eval_accuracy": 93.10000000000001,
|
| 235 |
+
"eval_average_metrics": 92.00454545454545,
|
| 236 |
+
"eval_f1": 90.90909090909089,
|
| 237 |
+
"eval_loss": 0.06345341354608536,
|
| 238 |
+
"eval_runtime": 4.6337,
|
| 239 |
+
"eval_samples_per_second": 215.808,
|
| 240 |
+
"step": 3800
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.35,
|
| 244 |
+
"learning_rate": 0.00026472352059264486,
|
| 245 |
+
"loss": 0.0745,
|
| 246 |
"step": 4000
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 0.35,
|
| 250 |
+
"eval_accuracy": 92.7,
|
| 251 |
+
"eval_average_metrics": 91.72389100126742,
|
| 252 |
+
"eval_f1": 90.74778200253485,
|
| 253 |
+
"eval_loss": 0.07112478464841843,
|
| 254 |
+
"eval_runtime": 4.5897,
|
| 255 |
+
"eval_samples_per_second": 217.881,
|
| 256 |
"step": 4000
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"epoch": 0.37,
|
| 260 |
+
"eval_accuracy": 92.4,
|
| 261 |
+
"eval_average_metrics": 91.38987341772153,
|
| 262 |
+
"eval_f1": 90.37974683544304,
|
| 263 |
+
"eval_loss": 0.06926184892654419,
|
| 264 |
+
"eval_runtime": 4.5334,
|
| 265 |
+
"eval_samples_per_second": 220.585,
|
| 266 |
+
"step": 4200
|
| 267 |
},
|
| 268 |
{
|
| 269 |
+
"epoch": 0.39,
|
| 270 |
+
"eval_accuracy": 94.19999999999999,
|
| 271 |
+
"eval_average_metrics": 93.2021505376344,
|
| 272 |
+
"eval_f1": 92.20430107526882,
|
| 273 |
+
"eval_loss": 0.06072888895869255,
|
| 274 |
+
"eval_runtime": 4.7545,
|
| 275 |
+
"eval_samples_per_second": 210.329,
|
| 276 |
+
"step": 4400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4,
|
| 280 |
+
"learning_rate": 0.00026031396066672545,
|
| 281 |
+
"loss": 0.0766,
|
| 282 |
"step": 4500
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 0.41,
|
| 286 |
+
"eval_accuracy": 93.89999999999999,
|
| 287 |
+
"eval_average_metrics": 92.8779038718291,
|
| 288 |
+
"eval_f1": 91.85580774365822,
|
| 289 |
+
"eval_loss": 0.06117413192987442,
|
| 290 |
+
"eval_runtime": 4.5554,
|
| 291 |
+
"eval_samples_per_second": 219.52,
|
| 292 |
+
"step": 4600
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.42,
|
| 296 |
+
"eval_accuracy": 94.0,
|
| 297 |
+
"eval_average_metrics": 93.13402061855669,
|
| 298 |
+
"eval_f1": 92.2680412371134,
|
| 299 |
+
"eval_loss": 0.060421667993068695,
|
| 300 |
+
"eval_runtime": 4.5444,
|
| 301 |
+
"eval_samples_per_second": 220.049,
|
| 302 |
+
"step": 4800
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.44,
|
| 306 |
+
"learning_rate": 0.00025590440074080604,
|
| 307 |
+
"loss": 0.0729,
|
| 308 |
"step": 5000
|
| 309 |
},
|
| 310 |
{
|
| 311 |
+
"epoch": 0.44,
|
| 312 |
+
"eval_accuracy": 93.0,
|
| 313 |
+
"eval_average_metrics": 91.95454545454547,
|
| 314 |
+
"eval_f1": 90.90909090909092,
|
| 315 |
+
"eval_loss": 0.06169410049915314,
|
| 316 |
+
"eval_runtime": 4.6688,
|
| 317 |
+
"eval_samples_per_second": 214.188,
|
| 318 |
"step": 5000
|
| 319 |
},
|
| 320 |
{
|
| 321 |
+
"epoch": 0.46,
|
| 322 |
+
"eval_accuracy": 94.3,
|
| 323 |
+
"eval_average_metrics": 93.3244966442953,
|
| 324 |
+
"eval_f1": 92.34899328859059,
|
| 325 |
+
"eval_loss": 0.0613214485347271,
|
| 326 |
+
"eval_runtime": 4.5049,
|
| 327 |
+
"eval_samples_per_second": 221.979,
|
| 328 |
+
"step": 5200
|
| 329 |
},
|
| 330 |
{
|
| 331 |
+
"epoch": 0.48,
|
| 332 |
+
"eval_accuracy": 93.7,
|
| 333 |
+
"eval_average_metrics": 92.71070959264125,
|
| 334 |
+
"eval_f1": 91.72141918528251,
|
| 335 |
+
"eval_loss": 0.06161003187298775,
|
| 336 |
+
"eval_runtime": 4.5685,
|
| 337 |
+
"eval_samples_per_second": 218.891,
|
| 338 |
+
"step": 5400
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.49,
|
| 342 |
+
"learning_rate": 0.0002514948408148867,
|
| 343 |
+
"loss": 0.071,
|
| 344 |
"step": 5500
|
| 345 |
},
|
| 346 |
{
|
| 347 |
+
"epoch": 0.49,
|
| 348 |
+
"eval_accuracy": 93.89999999999999,
|
| 349 |
+
"eval_average_metrics": 92.95262123197902,
|
| 350 |
+
"eval_f1": 92.00524246395806,
|
| 351 |
+
"eval_loss": 0.06118384748697281,
|
| 352 |
+
"eval_runtime": 4.564,
|
| 353 |
+
"eval_samples_per_second": 219.107,
|
| 354 |
+
"step": 5600
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.51,
|
| 358 |
+
"eval_accuracy": 93.0,
|
| 359 |
+
"eval_average_metrics": 92.0470737913486,
|
| 360 |
+
"eval_f1": 91.09414758269719,
|
| 361 |
+
"eval_loss": 0.06946446746587753,
|
| 362 |
+
"eval_runtime": 4.5787,
|
| 363 |
+
"eval_samples_per_second": 218.4,
|
| 364 |
+
"step": 5800
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 0.53,
|
| 368 |
+
"learning_rate": 0.0002470852808889673,
|
| 369 |
+
"loss": 0.0748,
|
| 370 |
"step": 6000
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"epoch": 0.53,
|
| 374 |
+
"eval_accuracy": 94.6,
|
| 375 |
+
"eval_average_metrics": 93.7095744680851,
|
| 376 |
+
"eval_f1": 92.81914893617021,
|
| 377 |
+
"eval_loss": 0.05765092372894287,
|
| 378 |
+
"eval_runtime": 4.5272,
|
| 379 |
+
"eval_samples_per_second": 220.889,
|
| 380 |
"step": 6000
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 0.55,
|
| 384 |
+
"eval_accuracy": 94.3,
|
| 385 |
+
"eval_average_metrics": 93.3244966442953,
|
| 386 |
+
"eval_f1": 92.34899328859059,
|
| 387 |
+
"eval_loss": 0.05894589051604271,
|
| 388 |
+
"eval_runtime": 4.6099,
|
| 389 |
+
"eval_samples_per_second": 216.924,
|
| 390 |
+
"step": 6200
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 0.56,
|
| 394 |
+
"eval_accuracy": 93.7,
|
| 395 |
+
"eval_average_metrics": 92.598987854251,
|
| 396 |
+
"eval_f1": 91.49797570850201,
|
| 397 |
+
"eval_loss": 0.061102479696273804,
|
| 398 |
+
"eval_runtime": 4.6948,
|
| 399 |
+
"eval_samples_per_second": 213.001,
|
| 400 |
+
"step": 6400
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.57,
|
| 404 |
+
"learning_rate": 0.00024267572096304786,
|
| 405 |
+
"loss": 0.074,
|
| 406 |
"step": 6500
|
| 407 |
},
|
| 408 |
{
|
| 409 |
+
"epoch": 0.58,
|
| 410 |
+
"eval_accuracy": 93.0,
|
| 411 |
+
"eval_average_metrics": 92.03571428571429,
|
| 412 |
+
"eval_f1": 91.07142857142858,
|
| 413 |
+
"eval_loss": 0.06452207267284393,
|
| 414 |
+
"eval_runtime": 4.6106,
|
| 415 |
+
"eval_samples_per_second": 216.891,
|
| 416 |
+
"step": 6600
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.6,
|
| 420 |
+
"eval_accuracy": 93.5,
|
| 421 |
+
"eval_average_metrics": 92.55645161290323,
|
| 422 |
+
"eval_f1": 91.61290322580645,
|
| 423 |
+
"eval_loss": 0.05938281863927841,
|
| 424 |
+
"eval_runtime": 4.5228,
|
| 425 |
+
"eval_samples_per_second": 221.102,
|
| 426 |
+
"step": 6800
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.62,
|
| 430 |
+
"learning_rate": 0.00023826616103712848,
|
| 431 |
+
"loss": 0.0738,
|
| 432 |
"step": 7000
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 0.62,
|
| 436 |
+
"eval_accuracy": 94.6,
|
| 437 |
+
"eval_average_metrics": 93.77519582245431,
|
| 438 |
+
"eval_f1": 92.95039164490862,
|
| 439 |
+
"eval_loss": 0.057858582586050034,
|
| 440 |
+
"eval_runtime": 4.5704,
|
| 441 |
+
"eval_samples_per_second": 218.797,
|
| 442 |
"step": 7000
|
| 443 |
},
|
| 444 |
{
|
| 445 |
+
"epoch": 0.63,
|
| 446 |
+
"eval_accuracy": 94.6,
|
| 447 |
+
"eval_average_metrics": 93.71909814323607,
|
| 448 |
+
"eval_f1": 92.83819628647215,
|
| 449 |
+
"eval_loss": 0.05671229586005211,
|
| 450 |
+
"eval_runtime": 4.4966,
|
| 451 |
+
"eval_samples_per_second": 222.39,
|
| 452 |
+
"step": 7200
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 0.65,
|
| 456 |
+
"eval_accuracy": 94.5,
|
| 457 |
+
"eval_average_metrics": 93.64580602883355,
|
| 458 |
+
"eval_f1": 92.7916120576671,
|
| 459 |
+
"eval_loss": 0.059491805732250214,
|
| 460 |
+
"eval_runtime": 4.5973,
|
| 461 |
+
"eval_samples_per_second": 217.521,
|
| 462 |
+
"step": 7400
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.66,
|
| 466 |
+
"learning_rate": 0.00023385660111120907,
|
| 467 |
+
"loss": 0.0746,
|
| 468 |
"step": 7500
|
| 469 |
},
|
| 470 |
{
|
| 471 |
+
"epoch": 0.67,
|
| 472 |
+
"eval_accuracy": 94.5,
|
| 473 |
+
"eval_average_metrics": 93.53879892037787,
|
| 474 |
+
"eval_f1": 92.57759784075573,
|
| 475 |
+
"eval_loss": 0.057486891746520996,
|
| 476 |
+
"eval_runtime": 4.6372,
|
| 477 |
+
"eval_samples_per_second": 215.649,
|
| 478 |
+
"step": 7600
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.69,
|
| 482 |
+
"eval_accuracy": 94.69999999999999,
|
| 483 |
+
"eval_average_metrics": 93.81194926568757,
|
| 484 |
+
"eval_f1": 92.92389853137516,
|
| 485 |
+
"eval_loss": 0.05628298968076706,
|
| 486 |
+
"eval_runtime": 4.6937,
|
| 487 |
+
"eval_samples_per_second": 213.051,
|
| 488 |
+
"step": 7800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.71,
|
| 492 |
+
"learning_rate": 0.0002294470411852897,
|
| 493 |
+
"loss": 0.0762,
|
| 494 |
"step": 8000
|
| 495 |
},
|
| 496 |
{
|
| 497 |
+
"epoch": 0.71,
|
| 498 |
+
"eval_accuracy": 94.39999999999999,
|
| 499 |
+
"eval_average_metrics": 93.4566844919786,
|
| 500 |
+
"eval_f1": 92.51336898395722,
|
| 501 |
+
"eval_loss": 0.05849047377705574,
|
| 502 |
+
"eval_runtime": 4.6139,
|
| 503 |
+
"eval_samples_per_second": 216.737,
|
| 504 |
"step": 8000
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 0.72,
|
| 508 |
+
"eval_accuracy": 94.69999999999999,
|
| 509 |
+
"eval_average_metrics": 93.86773981603153,
|
| 510 |
+
"eval_f1": 93.03547963206307,
|
| 511 |
+
"eval_loss": 0.056792281568050385,
|
| 512 |
+
"eval_runtime": 4.6916,
|
| 513 |
+
"eval_samples_per_second": 213.147,
|
| 514 |
+
"step": 8200
|
| 515 |
},
|
| 516 |
{
|
| 517 |
+
"epoch": 0.74,
|
| 518 |
+
"eval_accuracy": 94.5,
|
| 519 |
+
"eval_average_metrics": 93.59794156706508,
|
| 520 |
+
"eval_f1": 92.69588313413014,
|
| 521 |
+
"eval_loss": 0.05638590082526207,
|
| 522 |
+
"eval_runtime": 4.6952,
|
| 523 |
+
"eval_samples_per_second": 212.982,
|
| 524 |
+
"step": 8400
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.75,
|
| 528 |
+
"learning_rate": 0.0002250374812593703,
|
| 529 |
+
"loss": 0.0726,
|
| 530 |
"step": 8500
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"epoch": 0.76,
|
| 534 |
+
"eval_accuracy": 95.0,
|
| 535 |
+
"eval_average_metrics": 94.22774869109946,
|
| 536 |
+
"eval_f1": 93.45549738219894,
|
| 537 |
+
"eval_loss": 0.055720701813697815,
|
| 538 |
+
"eval_runtime": 4.5004,
|
| 539 |
+
"eval_samples_per_second": 222.204,
|
| 540 |
+
"step": 8600
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.78,
|
| 544 |
+
"eval_accuracy": 94.0,
|
| 545 |
+
"eval_average_metrics": 93.08355091383812,
|
| 546 |
+
"eval_f1": 92.16710182767625,
|
| 547 |
+
"eval_loss": 0.06084197014570236,
|
| 548 |
+
"eval_runtime": 4.5822,
|
| 549 |
+
"eval_samples_per_second": 218.238,
|
| 550 |
+
"step": 8800
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.79,
|
| 554 |
+
"learning_rate": 0.0002206279213334509,
|
| 555 |
+
"loss": 0.0734,
|
| 556 |
"step": 9000
|
| 557 |
},
|
| 558 |
{
|
| 559 |
+
"epoch": 0.79,
|
| 560 |
+
"eval_accuracy": 93.10000000000001,
|
| 561 |
+
"eval_average_metrics": 92.14386973180078,
|
| 562 |
+
"eval_f1": 91.18773946360155,
|
| 563 |
+
"eval_loss": 0.06530317664146423,
|
| 564 |
+
"eval_runtime": 4.5035,
|
| 565 |
+
"eval_samples_per_second": 222.05,
|
| 566 |
"step": 9000
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 0.81,
|
| 570 |
+
"eval_accuracy": 93.8,
|
| 571 |
+
"eval_average_metrics": 92.76666666666665,
|
| 572 |
+
"eval_f1": 91.73333333333332,
|
| 573 |
+
"eval_loss": 0.05946441367268562,
|
| 574 |
+
"eval_runtime": 4.8109,
|
| 575 |
+
"eval_samples_per_second": 207.861,
|
| 576 |
+
"step": 9200
|
| 577 |
},
|
| 578 |
{
|
| 579 |
+
"epoch": 0.83,
|
| 580 |
"eval_accuracy": 94.39999999999999,
|
| 581 |
+
"eval_average_metrics": 93.4566844919786,
|
| 582 |
+
"eval_f1": 92.51336898395722,
|
| 583 |
+
"eval_loss": 0.059339020401239395,
|
| 584 |
+
"eval_runtime": 4.5265,
|
| 585 |
+
"eval_samples_per_second": 220.922,
|
| 586 |
+
"step": 9400
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.84,
|
| 590 |
+
"learning_rate": 0.00021621836140753152,
|
| 591 |
+
"loss": 0.0731,
|
| 592 |
"step": 9500
|
| 593 |
},
|
| 594 |
{
|
| 595 |
+
"epoch": 0.85,
|
| 596 |
+
"eval_accuracy": 92.60000000000001,
|
| 597 |
+
"eval_average_metrics": 91.64005037783375,
|
| 598 |
+
"eval_f1": 90.6801007556675,
|
| 599 |
+
"eval_loss": 0.07186109572649002,
|
| 600 |
+
"eval_runtime": 4.5431,
|
| 601 |
+
"eval_samples_per_second": 220.114,
|
| 602 |
+
"step": 9600
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.86,
|
| 606 |
+
"eval_accuracy": 94.1,
|
| 607 |
+
"eval_average_metrics": 93.1633069828722,
|
| 608 |
+
"eval_f1": 92.2266139657444,
|
| 609 |
+
"eval_loss": 0.05946135148406029,
|
| 610 |
+
"eval_runtime": 4.5655,
|
| 611 |
+
"eval_samples_per_second": 219.036,
|
| 612 |
+
"step": 9800
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.88,
|
| 616 |
+
"learning_rate": 0.0002118088014816121,
|
| 617 |
+
"loss": 0.0733,
|
| 618 |
"step": 10000
|
| 619 |
},
|
| 620 |
{
|
| 621 |
+
"epoch": 0.88,
|
| 622 |
+
"eval_accuracy": 93.89999999999999,
|
| 623 |
+
"eval_average_metrics": 92.95262123197902,
|
| 624 |
+
"eval_f1": 92.00524246395806,
|
| 625 |
+
"eval_loss": 0.06076710671186447,
|
| 626 |
+
"eval_runtime": 4.5407,
|
| 627 |
+
"eval_samples_per_second": 220.229,
|
| 628 |
"step": 10000
|
| 629 |
},
|
| 630 |
{
|
| 631 |
+
"epoch": 0.9,
|
| 632 |
+
"eval_accuracy": 94.3,
|
| 633 |
+
"eval_average_metrics": 93.34492656875835,
|
| 634 |
+
"eval_f1": 92.3898531375167,
|
| 635 |
+
"eval_loss": 0.055939000099897385,
|
| 636 |
+
"eval_runtime": 4.4996,
|
| 637 |
+
"eval_samples_per_second": 222.24,
|
| 638 |
+
"step": 10200
|
| 639 |
},
|
| 640 |
{
|
| 641 |
+
"epoch": 0.92,
|
| 642 |
+
"eval_accuracy": 94.8,
|
| 643 |
+
"eval_average_metrics": 93.96084656084656,
|
| 644 |
+
"eval_f1": 93.12169312169313,
|
| 645 |
+
"eval_loss": 0.05636580288410187,
|
| 646 |
+
"eval_runtime": 4.4643,
|
| 647 |
+
"eval_samples_per_second": 223.998,
|
| 648 |
+
"step": 10400
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.93,
|
| 652 |
+
"learning_rate": 0.00020739924155569272,
|
| 653 |
+
"loss": 0.0738,
|
| 654 |
"step": 10500
|
| 655 |
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.93,
|
| 658 |
+
"eval_accuracy": 93.8,
|
| 659 |
+
"eval_average_metrics": 92.71081081081081,
|
| 660 |
+
"eval_f1": 91.62162162162161,
|
| 661 |
+
"eval_loss": 0.059175312519073486,
|
| 662 |
+
"eval_runtime": 4.7877,
|
| 663 |
+
"eval_samples_per_second": 208.867,
|
| 664 |
+
"step": 10600
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 0.95,
|
| 668 |
+
"eval_accuracy": 93.89999999999999,
|
| 669 |
+
"eval_average_metrics": 92.91026490066224,
|
| 670 |
+
"eval_f1": 91.9205298013245,
|
| 671 |
+
"eval_loss": 0.059644319117069244,
|
| 672 |
+
"eval_runtime": 4.6781,
|
| 673 |
+
"eval_samples_per_second": 213.761,
|
| 674 |
+
"step": 10800
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.97,
|
| 678 |
+
"learning_rate": 0.0002029896816297733,
|
| 679 |
+
"loss": 0.0752,
|
| 680 |
+
"step": 11000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.97,
|
| 684 |
+
"eval_accuracy": 93.60000000000001,
|
| 685 |
+
"eval_average_metrics": 92.622454308094,
|
| 686 |
+
"eval_f1": 91.644908616188,
|
| 687 |
+
"eval_loss": 0.061212606728076935,
|
| 688 |
+
"eval_runtime": 4.4874,
|
| 689 |
+
"eval_samples_per_second": 222.845,
|
| 690 |
+
"step": 11000
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.99,
|
| 694 |
+
"eval_accuracy": 94.3,
|
| 695 |
+
"eval_average_metrics": 93.41474442988203,
|
| 696 |
+
"eval_f1": 92.52948885976409,
|
| 697 |
+
"eval_loss": 0.059587035328149796,
|
| 698 |
+
"eval_runtime": 4.5784,
|
| 699 |
+
"eval_samples_per_second": 218.418,
|
| 700 |
+
"step": 11200
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 1.01,
|
| 704 |
+
"eval_accuracy": 94.1,
|
| 705 |
+
"eval_average_metrics": 93.2533462033462,
|
| 706 |
+
"eval_f1": 92.4066924066924,
|
| 707 |
+
"eval_loss": 0.060919877141714096,
|
| 708 |
+
"eval_runtime": 4.5416,
|
| 709 |
+
"eval_samples_per_second": 220.185,
|
| 710 |
+
"step": 11400
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.01,
|
| 714 |
+
"learning_rate": 0.00019858012170385393,
|
| 715 |
+
"loss": 0.0716,
|
| 716 |
+
"step": 11500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.02,
|
| 720 |
+
"eval_accuracy": 94.1,
|
| 721 |
+
"eval_average_metrics": 93.11141522029372,
|
| 722 |
+
"eval_f1": 92.12283044058745,
|
| 723 |
+
"eval_loss": 0.05961688980460167,
|
| 724 |
+
"eval_runtime": 4.5998,
|
| 725 |
+
"eval_samples_per_second": 217.402,
|
| 726 |
+
"step": 11600
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 1.04,
|
| 730 |
+
"eval_accuracy": 94.0,
|
| 731 |
+
"eval_average_metrics": 93.0212201591512,
|
| 732 |
+
"eval_f1": 92.04244031830238,
|
| 733 |
+
"eval_loss": 0.06122226640582085,
|
| 734 |
+
"eval_runtime": 4.6213,
|
| 735 |
+
"eval_samples_per_second": 216.391,
|
| 736 |
+
"step": 11800
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 1.06,
|
| 740 |
+
"learning_rate": 0.00019417056177793455,
|
| 741 |
+
"loss": 0.0713,
|
| 742 |
+
"step": 12000
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 1.06,
|
| 746 |
+
"eval_accuracy": 94.0,
|
| 747 |
+
"eval_average_metrics": 92.96774193548387,
|
| 748 |
+
"eval_f1": 91.93548387096774,
|
| 749 |
+
"eval_loss": 0.06119931861758232,
|
| 750 |
+
"eval_runtime": 4.5888,
|
| 751 |
+
"eval_samples_per_second": 217.92,
|
| 752 |
+
"step": 12000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.08,
|
| 756 |
+
"eval_accuracy": 94.3,
|
| 757 |
+
"eval_average_metrics": 93.31419919246298,
|
| 758 |
+
"eval_f1": 92.32839838492598,
|
| 759 |
+
"eval_loss": 0.05847727879881859,
|
| 760 |
+
"eval_runtime": 4.4633,
|
| 761 |
+
"eval_samples_per_second": 224.05,
|
| 762 |
+
"step": 12200
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 1.09,
|
| 766 |
+
"eval_accuracy": 93.4,
|
| 767 |
+
"eval_average_metrics": 92.39190600522193,
|
| 768 |
+
"eval_f1": 91.38381201044386,
|
| 769 |
+
"eval_loss": 0.06247144192457199,
|
| 770 |
+
"eval_runtime": 4.5667,
|
| 771 |
+
"eval_samples_per_second": 218.978,
|
| 772 |
+
"step": 12400
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.1,
|
| 776 |
+
"learning_rate": 0.00018976100185201514,
|
| 777 |
+
"loss": 0.0687,
|
| 778 |
+
"step": 12500
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 1.11,
|
| 782 |
+
"eval_accuracy": 93.89999999999999,
|
| 783 |
+
"eval_average_metrics": 93.04475032010242,
|
| 784 |
+
"eval_f1": 92.18950064020484,
|
| 785 |
+
"eval_loss": 0.0635332465171814,
|
| 786 |
+
"eval_runtime": 4.5944,
|
| 787 |
+
"eval_samples_per_second": 217.654,
|
| 788 |
+
"step": 12600
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.13,
|
| 792 |
+
"eval_accuracy": 94.0,
|
| 793 |
+
"eval_average_metrics": 93.05263157894737,
|
| 794 |
+
"eval_f1": 92.10526315789474,
|
| 795 |
+
"eval_loss": 0.06063272804021835,
|
| 796 |
+
"eval_runtime": 4.5058,
|
| 797 |
+
"eval_samples_per_second": 221.934,
|
| 798 |
+
"step": 12800
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 1.15,
|
| 802 |
+
"learning_rate": 0.00018535144192609576,
|
| 803 |
+
"loss": 0.0711,
|
| 804 |
+
"step": 13000
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 1.15,
|
| 808 |
+
"eval_accuracy": 94.0,
|
| 809 |
+
"eval_average_metrics": 93.12403100775194,
|
| 810 |
+
"eval_f1": 92.24806201550389,
|
| 811 |
+
"eval_loss": 0.06045162305235863,
|
| 812 |
+
"eval_runtime": 4.6598,
|
| 813 |
+
"eval_samples_per_second": 214.601,
|
| 814 |
+
"step": 13000
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.16,
|
| 818 |
+
"eval_accuracy": 94.39999999999999,
|
| 819 |
+
"eval_average_metrics": 93.50606860158311,
|
| 820 |
+
"eval_f1": 92.61213720316623,
|
| 821 |
+
"eval_loss": 0.06117108836770058,
|
| 822 |
+
"eval_runtime": 4.501,
|
| 823 |
+
"eval_samples_per_second": 222.172,
|
| 824 |
+
"step": 13200
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 1.18,
|
| 828 |
+
"eval_accuracy": 94.39999999999999,
|
| 829 |
+
"eval_average_metrics": 93.4962962962963,
|
| 830 |
+
"eval_f1": 92.5925925925926,
|
| 831 |
+
"eval_loss": 0.05846463143825531,
|
| 832 |
+
"eval_runtime": 4.4931,
|
| 833 |
+
"eval_samples_per_second": 222.561,
|
| 834 |
+
"step": 13400
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 1.19,
|
| 838 |
+
"learning_rate": 0.00018094188200017637,
|
| 839 |
+
"loss": 0.0694,
|
| 840 |
+
"step": 13500
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 1.2,
|
| 844 |
+
"eval_accuracy": 94.1,
|
| 845 |
+
"eval_average_metrics": 93.06889338731443,
|
| 846 |
+
"eval_f1": 92.03778677462888,
|
| 847 |
+
"eval_loss": 0.05917409434914589,
|
| 848 |
+
"eval_runtime": 4.6082,
|
| 849 |
+
"eval_samples_per_second": 217.003,
|
| 850 |
+
"step": 13600
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.22,
|
| 854 |
+
"eval_accuracy": 93.8,
|
| 855 |
+
"eval_average_metrics": 92.85300261096606,
|
| 856 |
+
"eval_f1": 91.9060052219321,
|
| 857 |
+
"eval_loss": 0.06280769407749176,
|
| 858 |
+
"eval_runtime": 4.5282,
|
| 859 |
+
"eval_samples_per_second": 220.836,
|
| 860 |
+
"step": 13800
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 1.23,
|
| 864 |
+
"learning_rate": 0.00017653232207425696,
|
| 865 |
+
"loss": 0.0741,
|
| 866 |
+
"step": 14000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 1.23,
|
| 870 |
+
"eval_accuracy": 93.60000000000001,
|
| 871 |
+
"eval_average_metrics": 92.65492227979274,
|
| 872 |
+
"eval_f1": 91.70984455958549,
|
| 873 |
+
"eval_loss": 0.06333824247121811,
|
| 874 |
+
"eval_runtime": 4.4743,
|
| 875 |
+
"eval_samples_per_second": 223.497,
|
| 876 |
+
"step": 14000
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.25,
|
| 880 |
+
"eval_accuracy": 93.89999999999999,
|
| 881 |
+
"eval_average_metrics": 92.94211563731932,
|
| 882 |
+
"eval_f1": 91.98423127463865,
|
| 883 |
+
"eval_loss": 0.06064913421869278,
|
| 884 |
+
"eval_runtime": 4.6765,
|
| 885 |
+
"eval_samples_per_second": 213.836,
|
| 886 |
+
"step": 14200
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 1.27,
|
| 890 |
+
"eval_accuracy": 92.60000000000001,
|
| 891 |
+
"eval_average_metrics": 91.62828282828283,
|
| 892 |
+
"eval_f1": 90.65656565656566,
|
| 893 |
+
"eval_loss": 0.07161322236061096,
|
| 894 |
+
"eval_runtime": 4.5138,
|
| 895 |
+
"eval_samples_per_second": 221.545,
|
| 896 |
+
"step": 14400
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 1.28,
|
| 900 |
+
"learning_rate": 0.00017212276214833758,
|
| 901 |
+
"loss": 0.0715,
|
| 902 |
+
"step": 14500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 1.29,
|
| 906 |
+
"eval_accuracy": 93.7,
|
| 907 |
+
"eval_average_metrics": 92.6998023715415,
|
| 908 |
+
"eval_f1": 91.699604743083,
|
| 909 |
+
"eval_loss": 0.06242042034864426,
|
| 910 |
+
"eval_runtime": 4.764,
|
| 911 |
+
"eval_samples_per_second": 209.909,
|
| 912 |
+
"step": 14600
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.31,
|
| 916 |
+
"eval_accuracy": 93.7,
|
| 917 |
+
"eval_average_metrics": 92.73235294117647,
|
| 918 |
+
"eval_f1": 91.76470588235294,
|
| 919 |
+
"eval_loss": 0.0626644566655159,
|
| 920 |
+
"eval_runtime": 4.4732,
|
| 921 |
+
"eval_samples_per_second": 223.555,
|
| 922 |
+
"step": 14800
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.32,
|
| 926 |
+
"learning_rate": 0.0001677132022224182,
|
| 927 |
+
"loss": 0.0714,
|
| 928 |
+
"step": 15000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 1.32,
|
| 932 |
+
"eval_accuracy": 94.39999999999999,
|
| 933 |
+
"eval_average_metrics": 93.54464751958224,
|
| 934 |
+
"eval_f1": 92.68929503916449,
|
| 935 |
+
"eval_loss": 0.05990656465291977,
|
| 936 |
+
"eval_runtime": 4.5922,
|
| 937 |
+
"eval_samples_per_second": 217.762,
|
| 938 |
+
"step": 15000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 1.34,
|
| 942 |
+
"eval_accuracy": 94.6,
|
| 943 |
+
"eval_average_metrics": 93.73799472295514,
|
| 944 |
+
"eval_f1": 92.87598944591029,
|
| 945 |
+
"eval_loss": 0.060957495123147964,
|
| 946 |
+
"eval_runtime": 4.4536,
|
| 947 |
+
"eval_samples_per_second": 224.537,
|
| 948 |
+
"step": 15200
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.36,
|
| 952 |
+
"eval_accuracy": 94.39999999999999,
|
| 953 |
+
"eval_average_metrics": 93.51578947368421,
|
| 954 |
+
"eval_f1": 92.63157894736842,
|
| 955 |
+
"eval_loss": 0.06167261675000191,
|
| 956 |
+
"eval_runtime": 4.4865,
|
| 957 |
+
"eval_samples_per_second": 222.89,
|
| 958 |
+
"step": 15400
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 1.37,
|
| 962 |
+
"learning_rate": 0.0001633036422964988,
|
| 963 |
+
"loss": 0.0707,
|
| 964 |
+
"step": 15500
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.38,
|
| 968 |
+
"eval_accuracy": 94.39999999999999,
|
| 969 |
+
"eval_average_metrics": 93.51578947368421,
|
| 970 |
+
"eval_f1": 92.63157894736842,
|
| 971 |
+
"eval_loss": 0.061066027730703354,
|
| 972 |
+
"eval_runtime": 4.5716,
|
| 973 |
+
"eval_samples_per_second": 218.74,
|
| 974 |
+
"step": 15600
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.39,
|
| 978 |
+
"eval_accuracy": 94.1,
|
| 979 |
+
"eval_average_metrics": 93.1633069828722,
|
| 980 |
+
"eval_f1": 92.2266139657444,
|
| 981 |
+
"eval_loss": 0.06235107034444809,
|
| 982 |
+
"eval_runtime": 4.499,
|
| 983 |
+
"eval_samples_per_second": 222.27,
|
| 984 |
+
"step": 15800
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 1.41,
|
| 988 |
+
"learning_rate": 0.00015889408237057938,
|
| 989 |
+
"loss": 0.0709,
|
| 990 |
+
"step": 16000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.41,
|
| 994 |
+
"eval_accuracy": 94.39999999999999,
|
| 995 |
+
"eval_average_metrics": 93.47659574468085,
|
| 996 |
+
"eval_f1": 92.55319148936171,
|
| 997 |
+
"eval_loss": 0.06194847822189331,
|
| 998 |
+
"eval_runtime": 4.5231,
|
| 999 |
+
"eval_samples_per_second": 221.086,
|
| 1000 |
+
"step": 16000
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 1.43,
|
| 1004 |
+
"eval_accuracy": 94.6,
|
| 1005 |
+
"eval_average_metrics": 93.76596858638743,
|
| 1006 |
+
"eval_f1": 92.93193717277488,
|
| 1007 |
+
"eval_loss": 0.05966123938560486,
|
| 1008 |
+
"eval_runtime": 4.637,
|
| 1009 |
+
"eval_samples_per_second": 215.656,
|
| 1010 |
+
"step": 16200
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.45,
|
| 1014 |
+
"eval_accuracy": 94.39999999999999,
|
| 1015 |
+
"eval_average_metrics": 93.44664879356569,
|
| 1016 |
+
"eval_f1": 92.49329758713138,
|
| 1017 |
+
"eval_loss": 0.06104936823248863,
|
| 1018 |
+
"eval_runtime": 4.5374,
|
| 1019 |
+
"eval_samples_per_second": 220.391,
|
| 1020 |
+
"step": 16400
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 1.46,
|
| 1024 |
+
"learning_rate": 0.00015448452244466002,
|
| 1025 |
+
"loss": 0.0729,
|
| 1026 |
+
"step": 16500
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 1.46,
|
| 1030 |
+
"eval_accuracy": 94.69999999999999,
|
| 1031 |
+
"eval_average_metrics": 93.87686762778506,
|
| 1032 |
+
"eval_f1": 93.05373525557013,
|
| 1033 |
+
"eval_loss": 0.06205834820866585,
|
| 1034 |
+
"eval_runtime": 4.4838,
|
| 1035 |
+
"eval_samples_per_second": 223.023,
|
| 1036 |
+
"step": 16600
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"epoch": 1.48,
|
| 1040 |
+
"eval_accuracy": 94.6,
|
| 1041 |
+
"eval_average_metrics": 93.78437500000001,
|
| 1042 |
+
"eval_f1": 92.96875000000001,
|
| 1043 |
+
"eval_loss": 0.06089754402637482,
|
| 1044 |
+
"eval_runtime": 4.5038,
|
| 1045 |
+
"eval_samples_per_second": 222.035,
|
| 1046 |
+
"step": 16800
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.5,
|
| 1050 |
+
"learning_rate": 0.00015007496251874061,
|
| 1051 |
+
"loss": 0.07,
|
| 1052 |
+
"step": 17000
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.5,
|
| 1056 |
+
"eval_accuracy": 94.19999999999999,
|
| 1057 |
+
"eval_average_metrics": 93.36288659793814,
|
| 1058 |
+
"eval_f1": 92.52577319587628,
|
| 1059 |
+
"eval_loss": 0.06112566590309143,
|
| 1060 |
+
"eval_runtime": 4.6026,
|
| 1061 |
+
"eval_samples_per_second": 217.269,
|
| 1062 |
+
"step": 17000
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.52,
|
| 1066 |
+
"eval_accuracy": 94.39999999999999,
|
| 1067 |
+
"eval_average_metrics": 93.48647214854111,
|
| 1068 |
+
"eval_f1": 92.57294429708222,
|
| 1069 |
+
"eval_loss": 0.06089947372674942,
|
| 1070 |
+
"eval_runtime": 4.5389,
|
| 1071 |
+
"eval_samples_per_second": 220.318,
|
| 1072 |
+
"step": 17200
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 1.53,
|
| 1076 |
+
"eval_accuracy": 94.1,
|
| 1077 |
+
"eval_average_metrics": 93.24354838709678,
|
| 1078 |
+
"eval_f1": 92.38709677419355,
|
| 1079 |
+
"eval_loss": 0.06110972911119461,
|
| 1080 |
+
"eval_runtime": 4.6859,
|
| 1081 |
+
"eval_samples_per_second": 213.405,
|
| 1082 |
+
"step": 17400
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 1.54,
|
| 1086 |
+
"learning_rate": 0.00014566540259282123,
|
| 1087 |
+
"loss": 0.0669,
|
| 1088 |
+
"step": 17500
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.55,
|
| 1092 |
+
"eval_accuracy": 94.3,
|
| 1093 |
+
"eval_average_metrics": 93.38513870541613,
|
| 1094 |
+
"eval_f1": 92.47027741083225,
|
| 1095 |
+
"eval_loss": 0.06174538657069206,
|
| 1096 |
+
"eval_runtime": 4.6674,
|
| 1097 |
+
"eval_samples_per_second": 214.254,
|
| 1098 |
+
"step": 17600
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 1.57,
|
| 1102 |
+
"eval_accuracy": 94.6,
|
| 1103 |
+
"eval_average_metrics": 93.7095744680851,
|
| 1104 |
+
"eval_f1": 92.81914893617021,
|
| 1105 |
+
"eval_loss": 0.059681929647922516,
|
| 1106 |
+
"eval_runtime": 4.6196,
|
| 1107 |
+
"eval_samples_per_second": 216.471,
|
| 1108 |
+
"step": 17800
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 1.59,
|
| 1112 |
+
"learning_rate": 0.00014125584266690182,
|
| 1113 |
+
"loss": 0.07,
|
| 1114 |
+
"step": 18000
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 1.59,
|
| 1118 |
+
"eval_accuracy": 94.19999999999999,
|
| 1119 |
+
"eval_average_metrics": 93.29422572178477,
|
| 1120 |
+
"eval_f1": 92.38845144356955,
|
| 1121 |
+
"eval_loss": 0.061346184462308884,
|
| 1122 |
+
"eval_runtime": 4.4984,
|
| 1123 |
+
"eval_samples_per_second": 222.301,
|
| 1124 |
+
"step": 18000
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 1.61,
|
| 1128 |
+
"eval_accuracy": 94.19999999999999,
|
| 1129 |
+
"eval_average_metrics": 93.28421052631577,
|
| 1130 |
+
"eval_f1": 92.36842105263158,
|
| 1131 |
+
"eval_loss": 0.06077203154563904,
|
| 1132 |
+
"eval_runtime": 4.454,
|
| 1133 |
+
"eval_samples_per_second": 224.518,
|
| 1134 |
+
"step": 18200
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 1.62,
|
| 1138 |
+
"eval_accuracy": 94.1,
|
| 1139 |
+
"eval_average_metrics": 93.13233731739707,
|
| 1140 |
+
"eval_f1": 92.16467463479415,
|
| 1141 |
+
"eval_loss": 0.05959217995405197,
|
| 1142 |
+
"eval_runtime": 4.588,
|
| 1143 |
+
"eval_samples_per_second": 217.959,
|
| 1144 |
+
"step": 18400
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.63,
|
| 1148 |
+
"learning_rate": 0.00013684628274098244,
|
| 1149 |
+
"loss": 0.069,
|
| 1150 |
+
"step": 18500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 1.64,
|
| 1154 |
+
"eval_accuracy": 94.39999999999999,
|
| 1155 |
+
"eval_average_metrics": 93.46666666666665,
|
| 1156 |
+
"eval_f1": 92.53333333333332,
|
| 1157 |
+
"eval_loss": 0.06017257645726204,
|
| 1158 |
+
"eval_runtime": 4.483,
|
| 1159 |
+
"eval_samples_per_second": 223.066,
|
| 1160 |
+
"step": 18600
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 1.66,
|
| 1164 |
+
"eval_accuracy": 94.5,
|
| 1165 |
+
"eval_average_metrics": 93.58821571238349,
|
| 1166 |
+
"eval_f1": 92.67643142476697,
|
| 1167 |
+
"eval_loss": 0.058851905167102814,
|
| 1168 |
+
"eval_runtime": 4.5048,
|
| 1169 |
+
"eval_samples_per_second": 221.985,
|
| 1170 |
+
"step": 18800
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 1.68,
|
| 1174 |
+
"learning_rate": 0.00013243672281506306,
|
| 1175 |
+
"loss": 0.0713,
|
| 1176 |
+
"step": 19000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 1.68,
|
| 1180 |
+
"eval_accuracy": 93.89999999999999,
|
| 1181 |
+
"eval_average_metrics": 92.97346805736636,
|
| 1182 |
+
"eval_f1": 92.04693611473273,
|
| 1183 |
+
"eval_loss": 0.06167756766080856,
|
| 1184 |
+
"eval_runtime": 4.4859,
|
| 1185 |
+
"eval_samples_per_second": 222.921,
|
| 1186 |
+
"step": 19000
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.69,
|
| 1190 |
+
"eval_accuracy": 93.7,
|
| 1191 |
+
"eval_average_metrics": 92.6998023715415,
|
| 1192 |
+
"eval_f1": 91.699604743083,
|
| 1193 |
+
"eval_loss": 0.06253690272569656,
|
| 1194 |
+
"eval_runtime": 4.582,
|
| 1195 |
+
"eval_samples_per_second": 218.244,
|
| 1196 |
+
"step": 19200
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 1.71,
|
| 1200 |
+
"eval_accuracy": 93.89999999999999,
|
| 1201 |
+
"eval_average_metrics": 93.00433376455368,
|
| 1202 |
+
"eval_f1": 92.10866752910736,
|
| 1203 |
+
"eval_loss": 0.06255872547626495,
|
| 1204 |
+
"eval_runtime": 4.5188,
|
| 1205 |
+
"eval_samples_per_second": 221.296,
|
| 1206 |
+
"step": 19400
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 1.72,
|
| 1210 |
+
"learning_rate": 0.00012802716288914365,
|
| 1211 |
+
"loss": 0.0699,
|
| 1212 |
+
"step": 19500
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 1.73,
|
| 1216 |
+
"eval_accuracy": 94.3,
|
| 1217 |
+
"eval_average_metrics": 93.40492772667542,
|
| 1218 |
+
"eval_f1": 92.50985545335085,
|
| 1219 |
+
"eval_loss": 0.062451381236314774,
|
| 1220 |
+
"eval_runtime": 4.5919,
|
| 1221 |
+
"eval_samples_per_second": 217.773,
|
| 1222 |
+
"step": 19600
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 1.75,
|
| 1226 |
+
"eval_accuracy": 94.0,
|
| 1227 |
+
"eval_average_metrics": 93.01063829787235,
|
| 1228 |
+
"eval_f1": 92.0212765957447,
|
| 1229 |
+
"eval_loss": 0.06319490820169449,
|
| 1230 |
+
"eval_runtime": 4.591,
|
| 1231 |
+
"eval_samples_per_second": 217.817,
|
| 1232 |
+
"step": 19800
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 1.76,
|
| 1236 |
+
"learning_rate": 0.00012361760296322426,
|
| 1237 |
+
"loss": 0.0698,
|
| 1238 |
+
"step": 20000
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 1.76,
|
| 1242 |
+
"eval_accuracy": 93.5,
|
| 1243 |
+
"eval_average_metrics": 92.51271186440678,
|
| 1244 |
+
"eval_f1": 91.52542372881356,
|
| 1245 |
+
"eval_loss": 0.06364640593528748,
|
| 1246 |
+
"eval_runtime": 4.5171,
|
| 1247 |
+
"eval_samples_per_second": 221.382,
|
| 1248 |
+
"step": 20000
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 1.78,
|
| 1252 |
+
"eval_accuracy": 93.89999999999999,
|
| 1253 |
+
"eval_average_metrics": 92.99409857328145,
|
| 1254 |
+
"eval_f1": 92.0881971465629,
|
| 1255 |
+
"eval_loss": 0.06635148823261261,
|
| 1256 |
+
"eval_runtime": 4.6206,
|
| 1257 |
+
"eval_samples_per_second": 216.422,
|
| 1258 |
+
"step": 20200
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 1.8,
|
| 1262 |
+
"eval_accuracy": 94.1,
|
| 1263 |
+
"eval_average_metrics": 93.11141522029372,
|
| 1264 |
+
"eval_f1": 92.12283044058745,
|
| 1265 |
+
"eval_loss": 0.0606299452483654,
|
| 1266 |
+
"eval_runtime": 4.4605,
|
| 1267 |
+
"eval_samples_per_second": 224.19,
|
| 1268 |
+
"step": 20400
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 1.81,
|
| 1272 |
+
"learning_rate": 0.00011920804303730487,
|
| 1273 |
+
"loss": 0.0703,
|
| 1274 |
+
"step": 20500
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 1.82,
|
| 1278 |
+
"eval_accuracy": 94.19999999999999,
|
| 1279 |
+
"eval_average_metrics": 93.32395833333332,
|
| 1280 |
+
"eval_f1": 92.44791666666666,
|
| 1281 |
+
"eval_loss": 0.060722097754478455,
|
| 1282 |
+
"eval_runtime": 4.5249,
|
| 1283 |
+
"eval_samples_per_second": 221.001,
|
| 1284 |
+
"step": 20600
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.83,
|
| 1288 |
+
"eval_accuracy": 93.8,
|
| 1289 |
+
"eval_average_metrics": 92.76666666666665,
|
| 1290 |
+
"eval_f1": 91.73333333333332,
|
| 1291 |
+
"eval_loss": 0.05862819775938988,
|
| 1292 |
+
"eval_runtime": 4.5187,
|
| 1293 |
+
"eval_samples_per_second": 221.304,
|
| 1294 |
+
"step": 20800
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 1.85,
|
| 1298 |
+
"learning_rate": 0.00011479848311138547,
|
| 1299 |
+
"loss": 0.0698,
|
| 1300 |
+
"step": 21000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 1.85,
|
| 1304 |
+
"eval_accuracy": 93.8,
|
| 1305 |
+
"eval_average_metrics": 92.79947089947089,
|
| 1306 |
+
"eval_f1": 91.7989417989418,
|
| 1307 |
+
"eval_loss": 0.06128830835223198,
|
| 1308 |
+
"eval_runtime": 4.521,
|
| 1309 |
+
"eval_samples_per_second": 221.188,
|
| 1310 |
+
"step": 21000
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 1.87,
|
| 1314 |
+
"eval_accuracy": 93.8,
|
| 1315 |
+
"eval_average_metrics": 92.87402597402597,
|
| 1316 |
+
"eval_f1": 91.94805194805194,
|
| 1317 |
+
"eval_loss": 0.06374780088663101,
|
| 1318 |
+
"eval_runtime": 4.4879,
|
| 1319 |
+
"eval_samples_per_second": 222.822,
|
| 1320 |
+
"step": 21200
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 1.89,
|
| 1324 |
+
"eval_accuracy": 93.89999999999999,
|
| 1325 |
+
"eval_average_metrics": 92.88874833555259,
|
| 1326 |
+
"eval_f1": 91.87749667110519,
|
| 1327 |
+
"eval_loss": 0.06154455617070198,
|
| 1328 |
+
"eval_runtime": 4.6011,
|
| 1329 |
+
"eval_samples_per_second": 217.337,
|
| 1330 |
+
"step": 21400
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 1.9,
|
| 1334 |
+
"learning_rate": 0.00011038892318546609,
|
| 1335 |
+
"loss": 0.0709,
|
| 1336 |
+
"step": 21500
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.9,
|
| 1340 |
+
"eval_accuracy": 94.5,
|
| 1341 |
+
"eval_average_metrics": 93.57843791722297,
|
| 1342 |
+
"eval_f1": 92.65687583444593,
|
| 1343 |
+
"eval_loss": 0.060043178498744965,
|
| 1344 |
+
"eval_runtime": 4.46,
|
| 1345 |
+
"eval_samples_per_second": 224.215,
|
| 1346 |
+
"step": 21600
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 1.92,
|
| 1350 |
+
"eval_accuracy": 93.89999999999999,
|
| 1351 |
+
"eval_average_metrics": 92.93155467720685,
|
| 1352 |
+
"eval_f1": 91.9631093544137,
|
| 1353 |
+
"eval_loss": 0.061132512986660004,
|
| 1354 |
+
"eval_runtime": 4.4987,
|
| 1355 |
+
"eval_samples_per_second": 222.287,
|
| 1356 |
+
"step": 21800
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 1.94,
|
| 1360 |
+
"learning_rate": 0.00010597936325954669,
|
| 1361 |
+
"loss": 0.0695,
|
| 1362 |
+
"step": 22000
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 1.94,
|
| 1366 |
+
"eval_accuracy": 93.5,
|
| 1367 |
+
"eval_average_metrics": 92.56724581724582,
|
| 1368 |
+
"eval_f1": 91.63449163449164,
|
| 1369 |
+
"eval_loss": 0.06395059078931808,
|
| 1370 |
+
"eval_runtime": 4.6548,
|
| 1371 |
+
"eval_samples_per_second": 214.832,
|
| 1372 |
+
"step": 22000
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 1.96,
|
| 1376 |
+
"eval_accuracy": 94.19999999999999,
|
| 1377 |
+
"eval_average_metrics": 93.29422572178477,
|
| 1378 |
+
"eval_f1": 92.38845144356955,
|
| 1379 |
+
"eval_loss": 0.06141780689358711,
|
| 1380 |
+
"eval_runtime": 4.4836,
|
| 1381 |
+
"eval_samples_per_second": 223.034,
|
| 1382 |
+
"step": 22200
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.98,
|
| 1386 |
+
"eval_accuracy": 94.5,
|
| 1387 |
+
"eval_average_metrics": 93.65522875816993,
|
| 1388 |
+
"eval_f1": 92.81045751633987,
|
| 1389 |
+
"eval_loss": 0.058759015053510666,
|
| 1390 |
+
"eval_runtime": 4.5162,
|
| 1391 |
+
"eval_samples_per_second": 221.426,
|
| 1392 |
+
"step": 22400
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 1.98,
|
| 1396 |
+
"learning_rate": 0.0001015698033336273,
|
| 1397 |
+
"loss": 0.0715,
|
| 1398 |
+
"step": 22500
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 1.99,
|
| 1402 |
+
"eval_accuracy": 93.89999999999999,
|
| 1403 |
+
"eval_average_metrics": 92.97346805736636,
|
| 1404 |
+
"eval_f1": 92.04693611473273,
|
| 1405 |
+
"eval_loss": 0.06228160858154297,
|
| 1406 |
+
"eval_runtime": 4.4726,
|
| 1407 |
+
"eval_samples_per_second": 223.582,
|
| 1408 |
+
"step": 22600
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 2.01,
|
| 1412 |
+
"eval_accuracy": 94.6,
|
| 1413 |
+
"eval_average_metrics": 93.73799472295514,
|
| 1414 |
+
"eval_f1": 92.87598944591029,
|
| 1415 |
+
"eval_loss": 0.05991463363170624,
|
| 1416 |
+
"eval_runtime": 4.5003,
|
| 1417 |
+
"eval_samples_per_second": 222.206,
|
| 1418 |
+
"step": 22800
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.03,
|
| 1422 |
+
"learning_rate": 9.71602434077079e-05,
|
| 1423 |
+
"loss": 0.0682,
|
| 1424 |
+
"step": 23000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.03,
|
| 1428 |
+
"eval_accuracy": 94.0,
|
| 1429 |
+
"eval_average_metrics": 93.1038961038961,
|
| 1430 |
+
"eval_f1": 92.20779220779221,
|
| 1431 |
+
"eval_loss": 0.061682794243097305,
|
| 1432 |
+
"eval_runtime": 4.611,
|
| 1433 |
+
"eval_samples_per_second": 216.874,
|
| 1434 |
+
"step": 23000
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 2.05,
|
| 1438 |
+
"eval_accuracy": 93.5,
|
| 1439 |
+
"eval_average_metrics": 92.55645161290323,
|
| 1440 |
+
"eval_f1": 91.61290322580645,
|
| 1441 |
+
"eval_loss": 0.06373216211795807,
|
| 1442 |
+
"eval_runtime": 4.6044,
|
| 1443 |
+
"eval_samples_per_second": 217.186,
|
| 1444 |
+
"step": 23200
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 2.06,
|
| 1448 |
+
"eval_accuracy": 94.3,
|
| 1449 |
+
"eval_average_metrics": 93.34492656875835,
|
| 1450 |
+
"eval_f1": 92.3898531375167,
|
| 1451 |
+
"eval_loss": 0.05869932472705841,
|
| 1452 |
+
"eval_runtime": 4.4706,
|
| 1453 |
+
"eval_samples_per_second": 223.684,
|
| 1454 |
+
"step": 23400
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 2.07,
|
| 1458 |
+
"learning_rate": 9.27506834817885e-05,
|
| 1459 |
+
"loss": 0.0652,
|
| 1460 |
+
"step": 23500
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 2.08,
|
| 1464 |
+
"eval_accuracy": 94.39999999999999,
|
| 1465 |
+
"eval_average_metrics": 93.50606860158311,
|
| 1466 |
+
"eval_f1": 92.61213720316623,
|
| 1467 |
+
"eval_loss": 0.06166525185108185,
|
| 1468 |
+
"eval_runtime": 4.5725,
|
| 1469 |
+
"eval_samples_per_second": 218.699,
|
| 1470 |
+
"step": 23600
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 2.1,
|
| 1474 |
+
"eval_accuracy": 94.19999999999999,
|
| 1475 |
+
"eval_average_metrics": 93.33376623376623,
|
| 1476 |
+
"eval_f1": 92.46753246753246,
|
| 1477 |
+
"eval_loss": 0.06055561453104019,
|
| 1478 |
+
"eval_runtime": 4.5492,
|
| 1479 |
+
"eval_samples_per_second": 219.818,
|
| 1480 |
+
"step": 23800
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 2.12,
|
| 1484 |
+
"learning_rate": 8.834112355586911e-05,
|
| 1485 |
+
"loss": 0.0691,
|
| 1486 |
+
"step": 24000
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 2.12,
|
| 1490 |
+
"eval_accuracy": 93.7,
|
| 1491 |
+
"eval_average_metrics": 92.75377113133939,
|
| 1492 |
+
"eval_f1": 91.8075422626788,
|
| 1493 |
+
"eval_loss": 0.06339309364557266,
|
| 1494 |
+
"eval_runtime": 4.6365,
|
| 1495 |
+
"eval_samples_per_second": 215.678,
|
| 1496 |
+
"step": 24000
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 2.13,
|
| 1500 |
+
"eval_accuracy": 94.1,
|
| 1501 |
+
"eval_average_metrics": 93.1633069828722,
|
| 1502 |
+
"eval_f1": 92.2266139657444,
|
| 1503 |
+
"eval_loss": 0.06319531798362732,
|
| 1504 |
+
"eval_runtime": 4.5722,
|
| 1505 |
+
"eval_samples_per_second": 218.712,
|
| 1506 |
+
"step": 24200
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 2.15,
|
| 1510 |
+
"eval_accuracy": 94.1,
|
| 1511 |
+
"eval_average_metrics": 93.14271523178809,
|
| 1512 |
+
"eval_f1": 92.18543046357617,
|
| 1513 |
+
"eval_loss": 0.060979247093200684,
|
| 1514 |
+
"eval_runtime": 4.4363,
|
| 1515 |
+
"eval_samples_per_second": 225.412,
|
| 1516 |
+
"step": 24400
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.16,
|
| 1520 |
+
"learning_rate": 8.393156362994973e-05,
|
| 1521 |
+
"loss": 0.0679,
|
| 1522 |
+
"step": 24500
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 2.17,
|
| 1526 |
+
"eval_accuracy": 94.3,
|
| 1527 |
+
"eval_average_metrics": 93.38513870541613,
|
| 1528 |
+
"eval_f1": 92.47027741083225,
|
| 1529 |
+
"eval_loss": 0.061841148883104324,
|
| 1530 |
+
"eval_runtime": 4.4945,
|
| 1531 |
+
"eval_samples_per_second": 222.493,
|
| 1532 |
+
"step": 24600
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 2.19,
|
| 1536 |
+
"eval_accuracy": 94.39999999999999,
|
| 1537 |
+
"eval_average_metrics": 93.51578947368421,
|
| 1538 |
+
"eval_f1": 92.63157894736842,
|
| 1539 |
+
"eval_loss": 0.06020021066069603,
|
| 1540 |
+
"eval_runtime": 4.6482,
|
| 1541 |
+
"eval_samples_per_second": 215.136,
|
| 1542 |
+
"step": 24800
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 2.2,
|
| 1546 |
+
"learning_rate": 7.952200370403033e-05,
|
| 1547 |
+
"loss": 0.0678,
|
| 1548 |
+
"step": 25000
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 2.2,
|
| 1552 |
+
"eval_accuracy": 94.69999999999999,
|
| 1553 |
+
"eval_average_metrics": 93.86773981603153,
|
| 1554 |
+
"eval_f1": 93.03547963206307,
|
| 1555 |
+
"eval_loss": 0.061626460403203964,
|
| 1556 |
+
"eval_runtime": 4.4847,
|
| 1557 |
+
"eval_samples_per_second": 222.982,
|
| 1558 |
+
"step": 25000
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 2.22,
|
| 1562 |
+
"eval_accuracy": 94.3,
|
| 1563 |
+
"eval_average_metrics": 93.35505992010653,
|
| 1564 |
+
"eval_f1": 92.41011984021304,
|
| 1565 |
+
"eval_loss": 0.05932234972715378,
|
| 1566 |
+
"eval_runtime": 4.444,
|
| 1567 |
+
"eval_samples_per_second": 225.02,
|
| 1568 |
+
"step": 25200
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 2.24,
|
| 1572 |
+
"eval_accuracy": 94.3,
|
| 1573 |
+
"eval_average_metrics": 93.35505992010653,
|
| 1574 |
+
"eval_f1": 92.41011984021304,
|
| 1575 |
+
"eval_loss": 0.05860959738492966,
|
| 1576 |
+
"eval_runtime": 4.4729,
|
| 1577 |
+
"eval_samples_per_second": 223.568,
|
| 1578 |
+
"step": 25400
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.25,
|
| 1582 |
+
"learning_rate": 7.511244377811093e-05,
|
| 1583 |
+
"loss": 0.0687,
|
| 1584 |
+
"step": 25500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 2.26,
|
| 1588 |
+
"eval_accuracy": 94.6,
|
| 1589 |
+
"eval_average_metrics": 93.74736842105261,
|
| 1590 |
+
"eval_f1": 92.89473684210525,
|
| 1591 |
+
"eval_loss": 0.05995591729879379,
|
| 1592 |
+
"eval_runtime": 4.6311,
|
| 1593 |
+
"eval_samples_per_second": 215.933,
|
| 1594 |
+
"step": 25600
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 2.28,
|
| 1598 |
+
"eval_accuracy": 94.39999999999999,
|
| 1599 |
+
"eval_average_metrics": 93.51578947368421,
|
| 1600 |
+
"eval_f1": 92.63157894736842,
|
| 1601 |
+
"eval_loss": 0.06067919358611107,
|
| 1602 |
+
"eval_runtime": 4.4705,
|
| 1603 |
+
"eval_samples_per_second": 223.69,
|
| 1604 |
+
"step": 25800
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 2.29,
|
| 1608 |
+
"learning_rate": 7.070288385219154e-05,
|
| 1609 |
+
"loss": 0.0665,
|
| 1610 |
+
"step": 26000
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 2.29,
|
| 1614 |
+
"eval_accuracy": 94.6,
|
| 1615 |
+
"eval_average_metrics": 93.74736842105261,
|
| 1616 |
+
"eval_f1": 92.89473684210525,
|
| 1617 |
+
"eval_loss": 0.06090604141354561,
|
| 1618 |
+
"eval_runtime": 4.4777,
|
| 1619 |
+
"eval_samples_per_second": 223.33,
|
| 1620 |
+
"step": 26000
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 2.31,
|
| 1624 |
+
"eval_accuracy": 94.5,
|
| 1625 |
+
"eval_average_metrics": 93.63633377135348,
|
| 1626 |
+
"eval_f1": 92.77266754270696,
|
| 1627 |
+
"eval_loss": 0.06175965070724487,
|
| 1628 |
+
"eval_runtime": 4.5456,
|
| 1629 |
+
"eval_samples_per_second": 219.993,
|
| 1630 |
+
"step": 26200
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 2.33,
|
| 1634 |
+
"eval_accuracy": 94.1,
|
| 1635 |
+
"eval_average_metrics": 93.1937908496732,
|
| 1636 |
+
"eval_f1": 92.2875816993464,
|
| 1637 |
+
"eval_loss": 0.062108419835567474,
|
| 1638 |
+
"eval_runtime": 4.5414,
|
| 1639 |
+
"eval_samples_per_second": 220.196,
|
| 1640 |
+
"step": 26400
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 2.34,
|
| 1644 |
+
"learning_rate": 6.629332392627216e-05,
|
| 1645 |
+
"loss": 0.0681,
|
| 1646 |
+
"step": 26500
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 2.35,
|
| 1650 |
+
"eval_accuracy": 94.39999999999999,
|
| 1651 |
+
"eval_average_metrics": 93.48647214854111,
|
| 1652 |
+
"eval_f1": 92.57294429708222,
|
| 1653 |
+
"eval_loss": 0.060741446912288666,
|
| 1654 |
+
"eval_runtime": 4.4624,
|
| 1655 |
+
"eval_samples_per_second": 224.096,
|
| 1656 |
+
"step": 26600
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 2.36,
|
| 1660 |
+
"eval_accuracy": 94.39999999999999,
|
| 1661 |
+
"eval_average_metrics": 93.4962962962963,
|
| 1662 |
+
"eval_f1": 92.5925925925926,
|
| 1663 |
+
"eval_loss": 0.06029416620731354,
|
| 1664 |
+
"eval_runtime": 4.5256,
|
| 1665 |
+
"eval_samples_per_second": 220.966,
|
| 1666 |
+
"step": 26800
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 2.38,
|
| 1670 |
+
"learning_rate": 6.188376400035276e-05,
|
| 1671 |
+
"loss": 0.0667,
|
| 1672 |
+
"step": 27000
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 2.38,
|
| 1676 |
+
"eval_accuracy": 94.69999999999999,
|
| 1677 |
+
"eval_average_metrics": 93.84933949801848,
|
| 1678 |
+
"eval_f1": 92.99867899603699,
|
| 1679 |
+
"eval_loss": 0.059210509061813354,
|
| 1680 |
+
"eval_runtime": 4.8741,
|
| 1681 |
+
"eval_samples_per_second": 205.167,
|
| 1682 |
+
"step": 27000
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 2.4,
|
| 1686 |
+
"eval_accuracy": 94.3,
|
| 1687 |
+
"eval_average_metrics": 93.41474442988203,
|
| 1688 |
+
"eval_f1": 92.52948885976409,
|
| 1689 |
+
"eval_loss": 0.0605180561542511,
|
| 1690 |
+
"eval_runtime": 4.5293,
|
| 1691 |
+
"eval_samples_per_second": 220.783,
|
| 1692 |
+
"step": 27200
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 2.42,
|
| 1696 |
+
"eval_accuracy": 94.5,
|
| 1697 |
+
"eval_average_metrics": 93.64580602883355,
|
| 1698 |
+
"eval_f1": 92.7916120576671,
|
| 1699 |
+
"eval_loss": 0.060811493545770645,
|
| 1700 |
+
"eval_runtime": 4.5424,
|
| 1701 |
+
"eval_samples_per_second": 220.147,
|
| 1702 |
+
"step": 27400
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 2.43,
|
| 1706 |
+
"learning_rate": 5.747420407443336e-05,
|
| 1707 |
+
"loss": 0.0685,
|
| 1708 |
+
"step": 27500
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 2.43,
|
| 1712 |
+
"eval_accuracy": 94.1,
|
| 1713 |
+
"eval_average_metrics": 93.11141522029372,
|
| 1714 |
+
"eval_f1": 92.12283044058745,
|
| 1715 |
+
"eval_loss": 0.05978462100028992,
|
| 1716 |
+
"eval_runtime": 4.4831,
|
| 1717 |
+
"eval_samples_per_second": 223.06,
|
| 1718 |
+
"step": 27600
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.45,
|
| 1722 |
+
"eval_accuracy": 93.8,
|
| 1723 |
+
"eval_average_metrics": 92.87402597402597,
|
| 1724 |
+
"eval_f1": 91.94805194805194,
|
| 1725 |
+
"eval_loss": 0.06267183274030685,
|
| 1726 |
+
"eval_runtime": 4.4576,
|
| 1727 |
+
"eval_samples_per_second": 224.334,
|
| 1728 |
+
"step": 27800
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 2.47,
|
| 1732 |
+
"learning_rate": 5.3064644148513973e-05,
|
| 1733 |
+
"loss": 0.0672,
|
| 1734 |
+
"step": 28000
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 2.47,
|
| 1738 |
+
"eval_accuracy": 94.0,
|
| 1739 |
+
"eval_average_metrics": 93.06299212598425,
|
| 1740 |
+
"eval_f1": 92.1259842519685,
|
| 1741 |
+
"eval_loss": 0.061355073004961014,
|
| 1742 |
+
"eval_runtime": 4.5194,
|
| 1743 |
+
"eval_samples_per_second": 221.27,
|
| 1744 |
+
"step": 28000
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 2.49,
|
| 1748 |
+
"eval_accuracy": 94.19999999999999,
|
| 1749 |
+
"eval_average_metrics": 93.29422572178477,
|
| 1750 |
+
"eval_f1": 92.38845144356955,
|
| 1751 |
+
"eval_loss": 0.06131287291646004,
|
| 1752 |
+
"eval_runtime": 4.5837,
|
| 1753 |
+
"eval_samples_per_second": 218.165,
|
| 1754 |
+
"step": 28200
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 2.5,
|
| 1758 |
+
"eval_accuracy": 94.0,
|
| 1759 |
+
"eval_average_metrics": 93.04221635883906,
|
| 1760 |
+
"eval_f1": 92.0844327176781,
|
| 1761 |
+
"eval_loss": 0.06105473265051842,
|
| 1762 |
+
"eval_runtime": 4.5101,
|
| 1763 |
+
"eval_samples_per_second": 221.726,
|
| 1764 |
+
"step": 28400
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 2.51,
|
| 1768 |
+
"learning_rate": 4.8655084222594584e-05,
|
| 1769 |
+
"loss": 0.0656,
|
| 1770 |
+
"step": 28500
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 2.52,
|
| 1774 |
+
"eval_accuracy": 94.19999999999999,
|
| 1775 |
+
"eval_average_metrics": 93.25384615384615,
|
| 1776 |
+
"eval_f1": 92.3076923076923,
|
| 1777 |
+
"eval_loss": 0.06093791127204895,
|
| 1778 |
+
"eval_runtime": 4.6588,
|
| 1779 |
+
"eval_samples_per_second": 214.647,
|
| 1780 |
+
"step": 28600
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 2.54,
|
| 1784 |
+
"eval_accuracy": 94.3,
|
| 1785 |
+
"eval_average_metrics": 93.40492772667542,
|
| 1786 |
+
"eval_f1": 92.50985545335085,
|
| 1787 |
+
"eval_loss": 0.061501096934080124,
|
| 1788 |
+
"eval_runtime": 4.5262,
|
| 1789 |
+
"eval_samples_per_second": 220.936,
|
| 1790 |
+
"step": 28800
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 2.56,
|
| 1794 |
+
"learning_rate": 4.424552429667519e-05,
|
| 1795 |
+
"loss": 0.067,
|
| 1796 |
+
"step": 29000
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 2.56,
|
| 1800 |
+
"eval_accuracy": 94.19999999999999,
|
| 1801 |
+
"eval_average_metrics": 93.27414248021108,
|
| 1802 |
+
"eval_f1": 92.34828496042218,
|
| 1803 |
+
"eval_loss": 0.05971948057413101,
|
| 1804 |
+
"eval_runtime": 4.5243,
|
| 1805 |
+
"eval_samples_per_second": 221.027,
|
| 1806 |
+
"step": 29000
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 2.58,
|
| 1810 |
+
"eval_accuracy": 93.60000000000001,
|
| 1811 |
+
"eval_average_metrics": 92.65492227979274,
|
| 1812 |
+
"eval_f1": 91.70984455958549,
|
| 1813 |
+
"eval_loss": 0.063376285135746,
|
| 1814 |
+
"eval_runtime": 4.6334,
|
| 1815 |
+
"eval_samples_per_second": 215.825,
|
| 1816 |
+
"step": 29200
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.59,
|
| 1820 |
+
"eval_accuracy": 94.19999999999999,
|
| 1821 |
+
"eval_average_metrics": 93.26402116402116,
|
| 1822 |
+
"eval_f1": 92.32804232804234,
|
| 1823 |
+
"eval_loss": 0.06081530451774597,
|
| 1824 |
+
"eval_runtime": 4.7045,
|
| 1825 |
+
"eval_samples_per_second": 212.561,
|
| 1826 |
+
"step": 29400
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 2.6,
|
| 1830 |
+
"learning_rate": 3.98359643707558e-05,
|
| 1831 |
+
"loss": 0.0675,
|
| 1832 |
+
"step": 29500
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 2.61,
|
| 1836 |
+
"eval_accuracy": 94.1,
|
| 1837 |
+
"eval_average_metrics": 93.18368283093054,
|
| 1838 |
+
"eval_f1": 92.26736566186108,
|
| 1839 |
+
"eval_loss": 0.062262628227472305,
|
| 1840 |
+
"eval_runtime": 4.6273,
|
| 1841 |
+
"eval_samples_per_second": 216.108,
|
| 1842 |
+
"step": 29600
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 2.63,
|
| 1846 |
+
"eval_accuracy": 94.3,
|
| 1847 |
+
"eval_average_metrics": 93.3751655629139,
|
| 1848 |
+
"eval_f1": 92.45033112582782,
|
| 1849 |
+
"eval_loss": 0.06007382273674011,
|
| 1850 |
+
"eval_runtime": 4.5698,
|
| 1851 |
+
"eval_samples_per_second": 218.83,
|
| 1852 |
+
"step": 29800
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 2.65,
|
| 1856 |
+
"learning_rate": 3.54264044448364e-05,
|
| 1857 |
+
"loss": 0.0682,
|
| 1858 |
+
"step": 30000
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.65,
|
| 1862 |
+
"eval_accuracy": 94.1,
|
| 1863 |
+
"eval_average_metrics": 93.1633069828722,
|
| 1864 |
+
"eval_f1": 92.2266139657444,
|
| 1865 |
+
"eval_loss": 0.0607917495071888,
|
| 1866 |
+
"eval_runtime": 4.6423,
|
| 1867 |
+
"eval_samples_per_second": 215.411,
|
| 1868 |
+
"step": 30000
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 2.66,
|
| 1872 |
+
"eval_accuracy": 94.39999999999999,
|
| 1873 |
+
"eval_average_metrics": 93.5254593175853,
|
| 1874 |
+
"eval_f1": 92.6509186351706,
|
| 1875 |
+
"eval_loss": 0.06171978637576103,
|
| 1876 |
+
"eval_runtime": 4.4956,
|
| 1877 |
+
"eval_samples_per_second": 222.439,
|
| 1878 |
+
"step": 30200
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 2.68,
|
| 1882 |
+
"eval_accuracy": 94.39999999999999,
|
| 1883 |
+
"eval_average_metrics": 93.4566844919786,
|
| 1884 |
+
"eval_f1": 92.51336898395722,
|
| 1885 |
+
"eval_loss": 0.05954898148775101,
|
| 1886 |
+
"eval_runtime": 4.5069,
|
| 1887 |
+
"eval_samples_per_second": 221.881,
|
| 1888 |
+
"step": 30400
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 2.69,
|
| 1892 |
+
"learning_rate": 3.1016844518917006e-05,
|
| 1893 |
+
"loss": 0.0684,
|
| 1894 |
+
"step": 30500
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 2.7,
|
| 1898 |
+
"eval_accuracy": 94.5,
|
| 1899 |
+
"eval_average_metrics": 93.64580602883355,
|
| 1900 |
+
"eval_f1": 92.7916120576671,
|
| 1901 |
+
"eval_loss": 0.06073066592216492,
|
| 1902 |
+
"eval_runtime": 4.5568,
|
| 1903 |
+
"eval_samples_per_second": 219.452,
|
| 1904 |
+
"step": 30600
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 2.72,
|
| 1908 |
+
"eval_accuracy": 94.5,
|
| 1909 |
+
"eval_average_metrics": 93.64580602883355,
|
| 1910 |
+
"eval_f1": 92.7916120576671,
|
| 1911 |
+
"eval_loss": 0.06212097778916359,
|
| 1912 |
+
"eval_runtime": 4.4991,
|
| 1913 |
+
"eval_samples_per_second": 222.265,
|
| 1914 |
+
"step": 30800
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.73,
|
| 1918 |
+
"learning_rate": 2.6607284592997617e-05,
|
| 1919 |
+
"loss": 0.0644,
|
| 1920 |
+
"step": 31000
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 2.73,
|
| 1924 |
+
"eval_accuracy": 94.5,
|
| 1925 |
+
"eval_average_metrics": 93.64580602883355,
|
| 1926 |
+
"eval_f1": 92.7916120576671,
|
| 1927 |
+
"eval_loss": 0.061464857310056686,
|
| 1928 |
+
"eval_runtime": 4.6313,
|
| 1929 |
+
"eval_samples_per_second": 215.924,
|
| 1930 |
+
"step": 31000
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 2.75,
|
| 1934 |
+
"eval_accuracy": 94.19999999999999,
|
| 1935 |
+
"eval_average_metrics": 93.28421052631577,
|
| 1936 |
+
"eval_f1": 92.36842105263158,
|
| 1937 |
+
"eval_loss": 0.06165764480829239,
|
| 1938 |
+
"eval_runtime": 4.4772,
|
| 1939 |
+
"eval_samples_per_second": 223.356,
|
| 1940 |
+
"step": 31200
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 2.77,
|
| 1944 |
+
"eval_accuracy": 94.19999999999999,
|
| 1945 |
+
"eval_average_metrics": 93.27414248021108,
|
| 1946 |
+
"eval_f1": 92.34828496042218,
|
| 1947 |
+
"eval_loss": 0.061222758144140244,
|
| 1948 |
+
"eval_runtime": 4.485,
|
| 1949 |
+
"eval_samples_per_second": 222.965,
|
| 1950 |
+
"step": 31400
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 2.78,
|
| 1954 |
+
"learning_rate": 2.219772466707822e-05,
|
| 1955 |
+
"loss": 0.0656,
|
| 1956 |
+
"step": 31500
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.79,
|
| 1960 |
+
"eval_accuracy": 94.3,
|
| 1961 |
+
"eval_average_metrics": 93.40492772667542,
|
| 1962 |
+
"eval_f1": 92.50985545335085,
|
| 1963 |
+
"eval_loss": 0.06175553798675537,
|
| 1964 |
+
"eval_runtime": 4.4473,
|
| 1965 |
+
"eval_samples_per_second": 224.857,
|
| 1966 |
+
"step": 31600
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 2.8,
|
| 1970 |
+
"eval_accuracy": 94.0,
|
| 1971 |
+
"eval_average_metrics": 93.04221635883906,
|
| 1972 |
+
"eval_f1": 92.0844327176781,
|
| 1973 |
+
"eval_loss": 0.06141304597258568,
|
| 1974 |
+
"eval_runtime": 4.5384,
|
| 1975 |
+
"eval_samples_per_second": 220.341,
|
| 1976 |
+
"step": 31800
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 2.82,
|
| 1980 |
+
"learning_rate": 1.778816474115883e-05,
|
| 1981 |
+
"loss": 0.0682,
|
| 1982 |
+
"step": 32000
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 2.82,
|
| 1986 |
+
"eval_accuracy": 94.3,
|
| 1987 |
+
"eval_average_metrics": 93.39505928853755,
|
| 1988 |
+
"eval_f1": 92.49011857707511,
|
| 1989 |
+
"eval_loss": 0.06122256815433502,
|
| 1990 |
+
"eval_runtime": 4.532,
|
| 1991 |
+
"eval_samples_per_second": 220.652,
|
| 1992 |
+
"step": 32000
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 2.84,
|
| 1996 |
+
"eval_accuracy": 94.39999999999999,
|
| 1997 |
+
"eval_average_metrics": 93.5254593175853,
|
| 1998 |
+
"eval_f1": 92.6509186351706,
|
| 1999 |
+
"eval_loss": 0.06179660186171532,
|
| 2000 |
+
"eval_runtime": 4.5432,
|
| 2001 |
+
"eval_samples_per_second": 220.11,
|
| 2002 |
+
"step": 32200
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 2.86,
|
| 2006 |
+
"eval_accuracy": 94.19999999999999,
|
| 2007 |
+
"eval_average_metrics": 93.26402116402116,
|
| 2008 |
+
"eval_f1": 92.32804232804234,
|
| 2009 |
+
"eval_loss": 0.060935478657484055,
|
| 2010 |
+
"eval_runtime": 4.5308,
|
| 2011 |
+
"eval_samples_per_second": 220.712,
|
| 2012 |
+
"step": 32400
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.87,
|
| 2016 |
+
"learning_rate": 1.3378604815239437e-05,
|
| 2017 |
+
"loss": 0.0628,
|
| 2018 |
+
"step": 32500
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 2.88,
|
| 2022 |
+
"eval_accuracy": 94.39999999999999,
|
| 2023 |
+
"eval_average_metrics": 93.51578947368421,
|
| 2024 |
+
"eval_f1": 92.63157894736842,
|
| 2025 |
+
"eval_loss": 0.06167520210146904,
|
| 2026 |
+
"eval_runtime": 4.5363,
|
| 2027 |
+
"eval_samples_per_second": 220.442,
|
| 2028 |
+
"step": 32600
|
| 2029 |
+
},
|
| 2030 |
+
{
|
| 2031 |
+
"epoch": 2.89,
|
| 2032 |
+
"eval_accuracy": 94.39999999999999,
|
| 2033 |
+
"eval_average_metrics": 93.51578947368421,
|
| 2034 |
+
"eval_f1": 92.63157894736842,
|
| 2035 |
+
"eval_loss": 0.061225228011608124,
|
| 2036 |
+
"eval_runtime": 4.5208,
|
| 2037 |
+
"eval_samples_per_second": 221.199,
|
| 2038 |
+
"step": 32800
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 2.91,
|
| 2042 |
+
"learning_rate": 8.969044889320046e-06,
|
| 2043 |
+
"loss": 0.0659,
|
| 2044 |
+
"step": 33000
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 2.91,
|
| 2048 |
+
"eval_accuracy": 94.3,
|
| 2049 |
+
"eval_average_metrics": 93.3751655629139,
|
| 2050 |
+
"eval_f1": 92.45033112582782,
|
| 2051 |
+
"eval_loss": 0.06039771810173988,
|
| 2052 |
+
"eval_runtime": 4.5643,
|
| 2053 |
+
"eval_samples_per_second": 219.093,
|
| 2054 |
+
"step": 33000
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.93,
|
| 2058 |
+
"eval_accuracy": 94.3,
|
| 2059 |
+
"eval_average_metrics": 93.40492772667542,
|
| 2060 |
+
"eval_f1": 92.50985545335085,
|
| 2061 |
+
"eval_loss": 0.06096240133047104,
|
| 2062 |
+
"eval_runtime": 4.5827,
|
| 2063 |
+
"eval_samples_per_second": 218.214,
|
| 2064 |
+
"step": 33200
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 2.95,
|
| 2068 |
+
"eval_accuracy": 94.3,
|
| 2069 |
+
"eval_average_metrics": 93.38513870541613,
|
| 2070 |
+
"eval_f1": 92.47027741083225,
|
| 2071 |
+
"eval_loss": 0.060673393309116364,
|
| 2072 |
+
"eval_runtime": 4.9126,
|
| 2073 |
+
"eval_samples_per_second": 203.559,
|
| 2074 |
+
"step": 33400
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.95,
|
| 2078 |
+
"learning_rate": 4.559484963400652e-06,
|
| 2079 |
+
"loss": 0.0692,
|
| 2080 |
+
"step": 33500
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 2.96,
|
| 2084 |
+
"eval_accuracy": 94.19999999999999,
|
| 2085 |
+
"eval_average_metrics": 93.26402116402116,
|
| 2086 |
+
"eval_f1": 92.32804232804234,
|
| 2087 |
+
"eval_loss": 0.06072871759533882,
|
| 2088 |
+
"eval_runtime": 4.5081,
|
| 2089 |
+
"eval_samples_per_second": 221.824,
|
| 2090 |
+
"step": 33600
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 2.98,
|
| 2094 |
+
"eval_accuracy": 94.19999999999999,
|
| 2095 |
+
"eval_average_metrics": 93.26402116402116,
|
| 2096 |
+
"eval_f1": 92.32804232804234,
|
| 2097 |
+
"eval_loss": 0.06088118627667427,
|
| 2098 |
+
"eval_runtime": 4.513,
|
| 2099 |
+
"eval_samples_per_second": 221.581,
|
| 2100 |
+
"step": 33800
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 3.0,
|
| 2104 |
+
"learning_rate": 1.4992503748125936e-07,
|
| 2105 |
+
"loss": 0.0654,
|
| 2106 |
+
"step": 34000
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 3.0,
|
| 2110 |
+
"eval_accuracy": 94.19999999999999,
|
| 2111 |
+
"eval_average_metrics": 93.26402116402116,
|
| 2112 |
+
"eval_f1": 92.32804232804234,
|
| 2113 |
+
"eval_loss": 0.060787323862314224,
|
| 2114 |
+
"eval_runtime": 4.526,
|
| 2115 |
+
"eval_samples_per_second": 220.947,
|
| 2116 |
+
"step": 34000
|
| 2117 |
+
},
|
| 2118 |
{
|
| 2119 |
"epoch": 3.0,
|
| 2120 |
+
"step": 34017,
|
| 2121 |
+
"total_flos": 1.0629344517601075e+17,
|
| 2122 |
+
"train_loss": 0.07169761398949699,
|
| 2123 |
+
"train_runtime": 13428.6442,
|
| 2124 |
+
"train_samples_per_second": 81.061,
|
| 2125 |
+
"train_steps_per_second": 2.533
|
| 2126 |
}
|
| 2127 |
],
|
| 2128 |
+
"max_steps": 34017,
|
| 2129 |
"num_train_epochs": 3,
|
| 2130 |
+
"total_flos": 1.0629344517601075e+17,
|
| 2131 |
"trial_name": null,
|
| 2132 |
"trial_params": null
|
| 2133 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3183
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7602f3c27193a11558b330a2f29b113c0d2a361010f05248a4c20f53c84c028c
|
| 3 |
size 3183
|
training_config.json
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
{"
|
|
|
|
| 1 |
+
{"dataset_config_name": ["en"], "delta_type": "lora", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "qqp", "eval_steps": 200, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "lora_r": 8, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 3, "output_dir": "outputs/bitfit/t5-base/qqp", "overwrite_output_dir": true, "per_device_eval_batch_size": 32, "per_device_train_batch_size": 32, "predict_with_generate": true, "push_to_hub": true, "save_steps": 200, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "split_validation_test": true, "task_name": "qqp", "test_dataset_config_name": ["en"], "test_dataset_name": "qqp", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas", "layer_norm", "final_layer_norm"], "warmup_steps": 0}
|