End of training
Browse files- README.md +3 -3
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/Aug20_17-01-35_big-desktop/events.out.tfevents.1724157760.big-desktop.242242.3 +3 -0
- train_results.json +6 -6
- trainer_state.json +1921 -103
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.7410179640718563
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.5745
|
| 36 |
+
- Accuracy: 0.7410
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 4.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos": 7.
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 29.820359281437124,
|
| 3 |
+
"eval_accuracy": 0.7410179640718563,
|
| 4 |
+
"eval_loss": 0.5744568705558777,
|
| 5 |
+
"eval_runtime": 4.0422,
|
| 6 |
+
"eval_samples_per_second": 165.256,
|
| 7 |
+
"eval_steps_per_second": 5.195,
|
| 8 |
+
"total_flos": 7.910788670992908e+18,
|
| 9 |
+
"train_loss": 0.5937201630159554,
|
| 10 |
+
"train_runtime": 3590.0203,
|
| 11 |
+
"train_samples_per_second": 89.181,
|
| 12 |
+
"train_steps_per_second": 0.694
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 4.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 29.820359281437124,
|
| 3 |
+
"eval_accuracy": 0.7410179640718563,
|
| 4 |
+
"eval_loss": 0.5744568705558777,
|
| 5 |
+
"eval_runtime": 4.0422,
|
| 6 |
+
"eval_samples_per_second": 165.256,
|
| 7 |
+
"eval_steps_per_second": 5.195
|
| 8 |
}
|
runs/Aug20_17-01-35_big-desktop/events.out.tfevents.1724157760.big-desktop.242242.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9eadb235120321d634d130a4dc36fdb8a2e93299678f03ae3a58169fa6d4fdc
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 7.
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 29.820359281437124,
|
| 3 |
+
"total_flos": 7.910788670992908e+18,
|
| 4 |
+
"train_loss": 0.5937201630159554,
|
| 5 |
+
"train_runtime": 3590.0203,
|
| 6 |
+
"train_samples_per_second": 89.181,
|
| 7 |
+
"train_steps_per_second": 0.694
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,222 +1,2040 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-rsna-2018/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.11976047904191617,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.23952095808383234,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.3592814371257485,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.47904191616766467,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.5988023952095808,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.718562874251497,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.8383233532934131,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.9580838323353293,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.9940119760479041,
|
| 69 |
-
"eval_accuracy": 0.
|
| 70 |
-
"eval_loss": 0.
|
| 71 |
-
"eval_runtime": 4.
|
| 72 |
-
"eval_samples_per_second":
|
| 73 |
-
"eval_steps_per_second": 4.
|
| 74 |
"step": 83
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 1.0778443113772456,
|
| 78 |
-
"grad_norm": 9.
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 1.1976047904191618,
|
| 85 |
-
"grad_norm": 8.
|
| 86 |
-
"learning_rate":
|
| 87 |
-
"loss": 0.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.3173652694610778,
|
| 92 |
-
"grad_norm": 6.
|
| 93 |
-
"learning_rate":
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.437125748502994,
|
| 99 |
-
"grad_norm":
|
| 100 |
-
"learning_rate": 2.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.55688622754491,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate": 2.
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.6766467065868262,
|
| 113 |
-
"grad_norm":
|
| 114 |
-
"learning_rate": 2.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 1.7964071856287425,
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.9161676646706587,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate":
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 2.0,
|
| 134 |
-
"eval_accuracy": 0.
|
| 135 |
-
"eval_loss": 0.
|
| 136 |
-
"eval_runtime": 4.
|
| 137 |
-
"eval_samples_per_second":
|
| 138 |
-
"eval_steps_per_second": 4.
|
| 139 |
"step": 167
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 2.035928143712575,
|
| 143 |
-
"grad_norm":
|
| 144 |
-
"learning_rate":
|
| 145 |
-
"loss": 0.
|
| 146 |
"step": 170
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 2.155688622754491,
|
| 150 |
-
"grad_norm":
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 2.2754491017964074,
|
| 157 |
-
"grad_norm":
|
| 158 |
-
"learning_rate":
|
| 159 |
-
"loss": 0.
|
| 160 |
"step": 190
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 2.3952095808383236,
|
| 164 |
-
"grad_norm":
|
| 165 |
-
"learning_rate":
|
| 166 |
-
"loss": 0.
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.5149700598802394,
|
| 171 |
-
"grad_norm": 4.
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 2.6347305389221556,
|
| 178 |
-
"grad_norm":
|
| 179 |
-
"learning_rate":
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 2.754491017964072,
|
| 185 |
-
"grad_norm": 4.
|
| 186 |
-
"learning_rate": 4.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 2.874251497005988,
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate":
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 2.
|
| 199 |
-
"
|
| 200 |
-
"
|
| 201 |
-
"
|
| 202 |
-
"
|
| 203 |
-
"eval_steps_per_second": 3.941,
|
| 204 |
-
"step": 249
|
| 205 |
},
|
| 206 |
{
|
| 207 |
-
"epoch": 2.
|
| 208 |
-
"
|
| 209 |
-
"
|
| 210 |
-
"
|
| 211 |
-
"
|
| 212 |
-
"
|
| 213 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 214 |
}
|
| 215 |
],
|
| 216 |
"logging_steps": 10,
|
| 217 |
-
"max_steps":
|
| 218 |
"num_input_tokens_seen": 0,
|
| 219 |
-
"num_train_epochs":
|
| 220 |
"save_steps": 500,
|
| 221 |
"stateful_callbacks": {
|
| 222 |
"TrainerControl": {
|
|
@@ -230,7 +2048,7 @@
|
|
| 230 |
"attributes": {}
|
| 231 |
}
|
| 232 |
},
|
| 233 |
-
"total_flos": 7.
|
| 234 |
"train_batch_size": 32,
|
| 235 |
"trial_name": null,
|
| 236 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.7410179640718563,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-rsna-2018/checkpoint-1002",
|
| 4 |
+
"epoch": 29.820359281437124,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 2490,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.11976047904191617,
|
| 13 |
+
"grad_norm": 5.50510835647583,
|
| 14 |
+
"learning_rate": 2.0080321285140564e-06,
|
| 15 |
+
"loss": 0.7371,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.23952095808383234,
|
| 20 |
+
"grad_norm": 10.76452350616455,
|
| 21 |
+
"learning_rate": 4.016064257028113e-06,
|
| 22 |
+
"loss": 0.7203,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.3592814371257485,
|
| 27 |
+
"grad_norm": 6.238582611083984,
|
| 28 |
+
"learning_rate": 6.024096385542169e-06,
|
| 29 |
+
"loss": 0.6998,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.47904191616766467,
|
| 34 |
+
"grad_norm": 6.8226189613342285,
|
| 35 |
+
"learning_rate": 8.032128514056226e-06,
|
| 36 |
+
"loss": 0.6751,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.5988023952095808,
|
| 41 |
+
"grad_norm": 8.670806884765625,
|
| 42 |
+
"learning_rate": 1.0040160642570281e-05,
|
| 43 |
+
"loss": 0.6305,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.718562874251497,
|
| 48 |
+
"grad_norm": 9.462418556213379,
|
| 49 |
+
"learning_rate": 1.2048192771084338e-05,
|
| 50 |
+
"loss": 0.6519,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.8383233532934131,
|
| 55 |
+
"grad_norm": 4.775774955749512,
|
| 56 |
+
"learning_rate": 1.4056224899598394e-05,
|
| 57 |
+
"loss": 0.6457,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.9580838323353293,
|
| 62 |
+
"grad_norm": 5.374617099761963,
|
| 63 |
+
"learning_rate": 1.606425702811245e-05,
|
| 64 |
+
"loss": 0.6448,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.9940119760479041,
|
| 69 |
+
"eval_accuracy": 0.6736526946107785,
|
| 70 |
+
"eval_loss": 0.6735118627548218,
|
| 71 |
+
"eval_runtime": 4.2254,
|
| 72 |
+
"eval_samples_per_second": 158.092,
|
| 73 |
+
"eval_steps_per_second": 4.97,
|
| 74 |
"step": 83
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 1.0778443113772456,
|
| 78 |
+
"grad_norm": 9.03003215789795,
|
| 79 |
+
"learning_rate": 1.8072289156626505e-05,
|
| 80 |
+
"loss": 0.7148,
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 1.1976047904191618,
|
| 85 |
+
"grad_norm": 8.153128623962402,
|
| 86 |
+
"learning_rate": 2.0080321285140562e-05,
|
| 87 |
+
"loss": 0.7236,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.3173652694610778,
|
| 92 |
+
"grad_norm": 6.323952674865723,
|
| 93 |
+
"learning_rate": 2.208835341365462e-05,
|
| 94 |
+
"loss": 0.7318,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.437125748502994,
|
| 99 |
+
"grad_norm": 8.060946464538574,
|
| 100 |
+
"learning_rate": 2.4096385542168677e-05,
|
| 101 |
+
"loss": 0.7131,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.55688622754491,
|
| 106 |
+
"grad_norm": 8.46826457977295,
|
| 107 |
+
"learning_rate": 2.6104417670682734e-05,
|
| 108 |
+
"loss": 0.7513,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 1.6766467065868262,
|
| 113 |
+
"grad_norm": 6.310857772827148,
|
| 114 |
+
"learning_rate": 2.8112449799196788e-05,
|
| 115 |
+
"loss": 0.7172,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 1.7964071856287425,
|
| 120 |
+
"grad_norm": 7.297194957733154,
|
| 121 |
+
"learning_rate": 3.012048192771085e-05,
|
| 122 |
+
"loss": 0.7141,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 1.9161676646706587,
|
| 127 |
+
"grad_norm": 5.612814903259277,
|
| 128 |
+
"learning_rate": 3.21285140562249e-05,
|
| 129 |
+
"loss": 0.736,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 2.0,
|
| 134 |
+
"eval_accuracy": 0.655688622754491,
|
| 135 |
+
"eval_loss": 0.6968684196472168,
|
| 136 |
+
"eval_runtime": 4.7501,
|
| 137 |
+
"eval_samples_per_second": 140.628,
|
| 138 |
+
"eval_steps_per_second": 4.421,
|
| 139 |
"step": 167
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 2.035928143712575,
|
| 143 |
+
"grad_norm": 6.799407482147217,
|
| 144 |
+
"learning_rate": 3.413654618473896e-05,
|
| 145 |
+
"loss": 0.7162,
|
| 146 |
"step": 170
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 2.155688622754491,
|
| 150 |
+
"grad_norm": 5.372466087341309,
|
| 151 |
+
"learning_rate": 3.614457831325301e-05,
|
| 152 |
+
"loss": 0.7359,
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 2.2754491017964074,
|
| 157 |
+
"grad_norm": 7.812543869018555,
|
| 158 |
+
"learning_rate": 3.815261044176707e-05,
|
| 159 |
+
"loss": 0.7076,
|
| 160 |
"step": 190
|
| 161 |
},
|
| 162 |
{
|
| 163 |
"epoch": 2.3952095808383236,
|
| 164 |
+
"grad_norm": 4.029458522796631,
|
| 165 |
+
"learning_rate": 4.0160642570281125e-05,
|
| 166 |
+
"loss": 0.7415,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 2.5149700598802394,
|
| 171 |
+
"grad_norm": 4.32454776763916,
|
| 172 |
+
"learning_rate": 4.2168674698795186e-05,
|
| 173 |
+
"loss": 0.721,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 2.6347305389221556,
|
| 178 |
+
"grad_norm": 4.972012042999268,
|
| 179 |
+
"learning_rate": 4.417670682730924e-05,
|
| 180 |
+
"loss": 0.7117,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 2.754491017964072,
|
| 185 |
+
"grad_norm": 4.34082555770874,
|
| 186 |
+
"learning_rate": 4.61847389558233e-05,
|
| 187 |
+
"loss": 0.7245,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 2.874251497005988,
|
| 192 |
+
"grad_norm": 6.3038787841796875,
|
| 193 |
+
"learning_rate": 4.8192771084337354e-05,
|
| 194 |
+
"loss": 0.7116,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 2.9940119760479043,
|
| 199 |
+
"grad_norm": 5.848024368286133,
|
| 200 |
+
"learning_rate": 4.9977688531905406e-05,
|
| 201 |
+
"loss": 0.6895,
|
| 202 |
+
"step": 250
|
|
|
|
|
|
|
| 203 |
},
|
| 204 |
{
|
| 205 |
+
"epoch": 2.9940119760479043,
|
| 206 |
+
"eval_accuracy": 0.6916167664670658,
|
| 207 |
+
"eval_loss": 0.6264948844909668,
|
| 208 |
+
"eval_runtime": 5.127,
|
| 209 |
+
"eval_samples_per_second": 130.29,
|
| 210 |
+
"eval_steps_per_second": 4.096,
|
| 211 |
+
"step": 250
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.1137724550898205,
|
| 215 |
+
"grad_norm": 5.597372055053711,
|
| 216 |
+
"learning_rate": 4.97545738509594e-05,
|
| 217 |
+
"loss": 0.6981,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.2335329341317367,
|
| 222 |
+
"grad_norm": 3.8474349975585938,
|
| 223 |
+
"learning_rate": 4.953145917001339e-05,
|
| 224 |
+
"loss": 0.6821,
|
| 225 |
+
"step": 270
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.3532934131736525,
|
| 229 |
+
"grad_norm": 4.591520309448242,
|
| 230 |
+
"learning_rate": 4.930834448906738e-05,
|
| 231 |
+
"loss": 0.6935,
|
| 232 |
+
"step": 280
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.4730538922155687,
|
| 236 |
+
"grad_norm": 7.39481258392334,
|
| 237 |
+
"learning_rate": 4.908522980812137e-05,
|
| 238 |
+
"loss": 0.7123,
|
| 239 |
+
"step": 290
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.592814371257485,
|
| 243 |
+
"grad_norm": 4.609263896942139,
|
| 244 |
+
"learning_rate": 4.886211512717537e-05,
|
| 245 |
+
"loss": 0.7209,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.712574850299401,
|
| 250 |
+
"grad_norm": 6.359745502471924,
|
| 251 |
+
"learning_rate": 4.8639000446229364e-05,
|
| 252 |
+
"loss": 0.6891,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.8323353293413174,
|
| 257 |
+
"grad_norm": 3.9733974933624268,
|
| 258 |
+
"learning_rate": 4.8415885765283355e-05,
|
| 259 |
+
"loss": 0.6796,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.9520958083832336,
|
| 264 |
+
"grad_norm": 5.370658874511719,
|
| 265 |
+
"learning_rate": 4.8192771084337354e-05,
|
| 266 |
+
"loss": 0.6631,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.0,
|
| 271 |
+
"eval_accuracy": 0.7155688622754491,
|
| 272 |
+
"eval_loss": 0.627507209777832,
|
| 273 |
+
"eval_runtime": 5.0645,
|
| 274 |
+
"eval_samples_per_second": 131.899,
|
| 275 |
+
"eval_steps_per_second": 4.147,
|
| 276 |
+
"step": 334
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.07185628742515,
|
| 280 |
+
"grad_norm": 4.173786163330078,
|
| 281 |
+
"learning_rate": 4.7969656403391346e-05,
|
| 282 |
+
"loss": 0.6847,
|
| 283 |
+
"step": 340
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 4.191616766467066,
|
| 287 |
+
"grad_norm": 8.743853569030762,
|
| 288 |
+
"learning_rate": 4.774654172244534e-05,
|
| 289 |
+
"loss": 0.6653,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.311377245508982,
|
| 294 |
+
"grad_norm": 3.8112645149230957,
|
| 295 |
+
"learning_rate": 4.7523427041499336e-05,
|
| 296 |
+
"loss": 0.682,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.431137724550898,
|
| 301 |
+
"grad_norm": 4.91067361831665,
|
| 302 |
+
"learning_rate": 4.730031236055333e-05,
|
| 303 |
+
"loss": 0.6849,
|
| 304 |
+
"step": 370
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.550898203592815,
|
| 308 |
+
"grad_norm": 7.876720428466797,
|
| 309 |
+
"learning_rate": 4.707719767960732e-05,
|
| 310 |
+
"loss": 0.665,
|
| 311 |
+
"step": 380
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 4.6706586826347305,
|
| 315 |
+
"grad_norm": 3.3110787868499756,
|
| 316 |
+
"learning_rate": 4.685408299866131e-05,
|
| 317 |
+
"loss": 0.6923,
|
| 318 |
+
"step": 390
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.790419161676647,
|
| 322 |
+
"grad_norm": 4.038461208343506,
|
| 323 |
+
"learning_rate": 4.663096831771531e-05,
|
| 324 |
+
"loss": 0.6776,
|
| 325 |
+
"step": 400
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 4.910179640718563,
|
| 329 |
+
"grad_norm": 4.028420448303223,
|
| 330 |
+
"learning_rate": 4.64078536367693e-05,
|
| 331 |
+
"loss": 0.6725,
|
| 332 |
+
"step": 410
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 4.994011976047904,
|
| 336 |
+
"eval_accuracy": 0.7125748502994012,
|
| 337 |
+
"eval_loss": 0.6311057806015015,
|
| 338 |
+
"eval_runtime": 5.1897,
|
| 339 |
+
"eval_samples_per_second": 128.717,
|
| 340 |
+
"eval_steps_per_second": 4.046,
|
| 341 |
+
"step": 417
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 5.029940119760479,
|
| 345 |
+
"grad_norm": 4.9200215339660645,
|
| 346 |
+
"learning_rate": 4.61847389558233e-05,
|
| 347 |
+
"loss": 0.6848,
|
| 348 |
+
"step": 420
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.149700598802395,
|
| 352 |
+
"grad_norm": 11.384115219116211,
|
| 353 |
+
"learning_rate": 4.596162427487729e-05,
|
| 354 |
+
"loss": 0.6651,
|
| 355 |
+
"step": 430
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 5.269461077844311,
|
| 359 |
+
"grad_norm": 4.320120811462402,
|
| 360 |
+
"learning_rate": 4.5738509593931284e-05,
|
| 361 |
+
"loss": 0.6627,
|
| 362 |
+
"step": 440
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 5.389221556886228,
|
| 366 |
+
"grad_norm": 3.2249197959899902,
|
| 367 |
+
"learning_rate": 4.5515394912985275e-05,
|
| 368 |
+
"loss": 0.695,
|
| 369 |
+
"step": 450
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.508982035928144,
|
| 373 |
+
"grad_norm": 6.424835681915283,
|
| 374 |
+
"learning_rate": 4.529228023203927e-05,
|
| 375 |
+
"loss": 0.665,
|
| 376 |
+
"step": 460
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 5.62874251497006,
|
| 380 |
+
"grad_norm": 3.735926389694214,
|
| 381 |
+
"learning_rate": 4.506916555109326e-05,
|
| 382 |
+
"loss": 0.6484,
|
| 383 |
+
"step": 470
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 5.748502994011976,
|
| 387 |
+
"grad_norm": 6.180431365966797,
|
| 388 |
+
"learning_rate": 4.484605087014726e-05,
|
| 389 |
+
"loss": 0.6623,
|
| 390 |
+
"step": 480
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.868263473053892,
|
| 394 |
+
"grad_norm": 4.55112886428833,
|
| 395 |
+
"learning_rate": 4.4622936189201256e-05,
|
| 396 |
+
"loss": 0.6954,
|
| 397 |
+
"step": 490
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 5.9880239520958085,
|
| 401 |
+
"grad_norm": 5.922323226928711,
|
| 402 |
+
"learning_rate": 4.439982150825525e-05,
|
| 403 |
+
"loss": 0.6778,
|
| 404 |
+
"step": 500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 6.0,
|
| 408 |
+
"eval_accuracy": 0.7065868263473054,
|
| 409 |
+
"eval_loss": 0.619443953037262,
|
| 410 |
+
"eval_runtime": 4.9491,
|
| 411 |
+
"eval_samples_per_second": 134.974,
|
| 412 |
+
"eval_steps_per_second": 4.243,
|
| 413 |
+
"step": 501
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.107784431137724,
|
| 417 |
+
"grad_norm": 4.259535789489746,
|
| 418 |
+
"learning_rate": 4.417670682730924e-05,
|
| 419 |
+
"loss": 0.6365,
|
| 420 |
+
"step": 510
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 6.227544910179641,
|
| 424 |
+
"grad_norm": 3.808413028717041,
|
| 425 |
+
"learning_rate": 4.395359214636323e-05,
|
| 426 |
+
"loss": 0.6913,
|
| 427 |
+
"step": 520
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 6.347305389221557,
|
| 431 |
+
"grad_norm": 4.0178632736206055,
|
| 432 |
+
"learning_rate": 4.373047746541722e-05,
|
| 433 |
+
"loss": 0.7112,
|
| 434 |
+
"step": 530
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.467065868263473,
|
| 438 |
+
"grad_norm": 3.5464377403259277,
|
| 439 |
+
"learning_rate": 4.350736278447122e-05,
|
| 440 |
+
"loss": 0.6287,
|
| 441 |
+
"step": 540
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.586826347305389,
|
| 445 |
+
"grad_norm": 5.3495612144470215,
|
| 446 |
+
"learning_rate": 4.328424810352521e-05,
|
| 447 |
+
"loss": 0.6441,
|
| 448 |
+
"step": 550
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.706586826347305,
|
| 452 |
+
"grad_norm": 3.6895763874053955,
|
| 453 |
+
"learning_rate": 4.306113342257921e-05,
|
| 454 |
+
"loss": 0.6492,
|
| 455 |
+
"step": 560
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.826347305389222,
|
| 459 |
+
"grad_norm": 6.391328811645508,
|
| 460 |
+
"learning_rate": 4.2838018741633203e-05,
|
| 461 |
+
"loss": 0.6347,
|
| 462 |
+
"step": 570
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 6.946107784431137,
|
| 466 |
+
"grad_norm": 3.929858922958374,
|
| 467 |
+
"learning_rate": 4.2614904060687195e-05,
|
| 468 |
+
"loss": 0.6734,
|
| 469 |
+
"step": 580
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 6.994011976047904,
|
| 473 |
+
"eval_accuracy": 0.7140718562874252,
|
| 474 |
+
"eval_loss": 0.602377712726593,
|
| 475 |
+
"eval_runtime": 5.4236,
|
| 476 |
+
"eval_samples_per_second": 123.166,
|
| 477 |
+
"eval_steps_per_second": 3.872,
|
| 478 |
+
"step": 584
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 7.065868263473054,
|
| 482 |
+
"grad_norm": 3.6292643547058105,
|
| 483 |
+
"learning_rate": 4.239178937974119e-05,
|
| 484 |
+
"loss": 0.6651,
|
| 485 |
+
"step": 590
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.18562874251497,
|
| 489 |
+
"grad_norm": 5.194599628448486,
|
| 490 |
+
"learning_rate": 4.2168674698795186e-05,
|
| 491 |
+
"loss": 0.643,
|
| 492 |
+
"step": 600
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.3053892215568865,
|
| 496 |
+
"grad_norm": 4.0095953941345215,
|
| 497 |
+
"learning_rate": 4.194556001784918e-05,
|
| 498 |
+
"loss": 0.6436,
|
| 499 |
+
"step": 610
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.425149700598802,
|
| 503 |
+
"grad_norm": 4.267141819000244,
|
| 504 |
+
"learning_rate": 4.172244533690317e-05,
|
| 505 |
+
"loss": 0.6114,
|
| 506 |
+
"step": 620
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.544910179640718,
|
| 510 |
+
"grad_norm": 3.5301904678344727,
|
| 511 |
+
"learning_rate": 4.149933065595716e-05,
|
| 512 |
+
"loss": 0.6326,
|
| 513 |
+
"step": 630
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.664670658682635,
|
| 517 |
+
"grad_norm": 3.8866894245147705,
|
| 518 |
+
"learning_rate": 4.127621597501116e-05,
|
| 519 |
+
"loss": 0.6331,
|
| 520 |
+
"step": 640
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.7844311377245505,
|
| 524 |
+
"grad_norm": 3.902667284011841,
|
| 525 |
+
"learning_rate": 4.105310129406515e-05,
|
| 526 |
+
"loss": 0.6515,
|
| 527 |
+
"step": 650
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.904191616766467,
|
| 531 |
+
"grad_norm": 4.829390525817871,
|
| 532 |
+
"learning_rate": 4.082998661311915e-05,
|
| 533 |
+
"loss": 0.6231,
|
| 534 |
+
"step": 660
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 8.0,
|
| 538 |
+
"eval_accuracy": 0.7230538922155688,
|
| 539 |
+
"eval_loss": 0.6081866025924683,
|
| 540 |
+
"eval_runtime": 5.7887,
|
| 541 |
+
"eval_samples_per_second": 115.397,
|
| 542 |
+
"eval_steps_per_second": 3.628,
|
| 543 |
+
"step": 668
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 8.023952095808383,
|
| 547 |
+
"grad_norm": 4.755038261413574,
|
| 548 |
+
"learning_rate": 4.060687193217314e-05,
|
| 549 |
+
"loss": 0.6261,
|
| 550 |
+
"step": 670
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 8.1437125748503,
|
| 554 |
+
"grad_norm": 3.4586455821990967,
|
| 555 |
+
"learning_rate": 4.038375725122713e-05,
|
| 556 |
+
"loss": 0.6534,
|
| 557 |
+
"step": 680
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 8.263473053892216,
|
| 561 |
+
"grad_norm": 8.194857597351074,
|
| 562 |
+
"learning_rate": 4.0160642570281125e-05,
|
| 563 |
+
"loss": 0.6329,
|
| 564 |
+
"step": 690
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 8.383233532934131,
|
| 568 |
+
"grad_norm": 2.9734175205230713,
|
| 569 |
+
"learning_rate": 3.993752788933512e-05,
|
| 570 |
+
"loss": 0.6676,
|
| 571 |
+
"step": 700
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 8.502994011976048,
|
| 575 |
+
"grad_norm": 5.66069221496582,
|
| 576 |
+
"learning_rate": 3.9714413208389115e-05,
|
| 577 |
+
"loss": 0.6178,
|
| 578 |
+
"step": 710
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 8.622754491017965,
|
| 582 |
+
"grad_norm": 3.4166412353515625,
|
| 583 |
+
"learning_rate": 3.949129852744311e-05,
|
| 584 |
+
"loss": 0.6399,
|
| 585 |
+
"step": 720
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.74251497005988,
|
| 589 |
+
"grad_norm": 5.076518535614014,
|
| 590 |
+
"learning_rate": 3.9268183846497105e-05,
|
| 591 |
+
"loss": 0.6163,
|
| 592 |
+
"step": 730
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 8.862275449101796,
|
| 596 |
+
"grad_norm": 3.32446551322937,
|
| 597 |
+
"learning_rate": 3.90450691655511e-05,
|
| 598 |
+
"loss": 0.6152,
|
| 599 |
+
"step": 740
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 8.982035928143713,
|
| 603 |
+
"grad_norm": 4.711836338043213,
|
| 604 |
+
"learning_rate": 3.882195448460509e-05,
|
| 605 |
+
"loss": 0.6164,
|
| 606 |
+
"step": 750
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 8.994011976047904,
|
| 610 |
+
"eval_accuracy": 0.7170658682634731,
|
| 611 |
+
"eval_loss": 0.5845786333084106,
|
| 612 |
+
"eval_runtime": 4.984,
|
| 613 |
+
"eval_samples_per_second": 134.029,
|
| 614 |
+
"eval_steps_per_second": 4.213,
|
| 615 |
+
"step": 751
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 9.10179640718563,
|
| 619 |
+
"grad_norm": 4.9026055335998535,
|
| 620 |
+
"learning_rate": 3.859883980365908e-05,
|
| 621 |
+
"loss": 0.623,
|
| 622 |
+
"step": 760
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 9.221556886227544,
|
| 626 |
+
"grad_norm": 5.39790678024292,
|
| 627 |
+
"learning_rate": 3.837572512271307e-05,
|
| 628 |
+
"loss": 0.5958,
|
| 629 |
+
"step": 770
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 9.341317365269461,
|
| 633 |
+
"grad_norm": 4.951222896575928,
|
| 634 |
+
"learning_rate": 3.815261044176707e-05,
|
| 635 |
+
"loss": 0.6206,
|
| 636 |
+
"step": 780
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 9.461077844311378,
|
| 640 |
+
"grad_norm": 3.8360514640808105,
|
| 641 |
+
"learning_rate": 3.792949576082106e-05,
|
| 642 |
+
"loss": 0.638,
|
| 643 |
+
"step": 790
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 9.580838323353294,
|
| 647 |
+
"grad_norm": 4.05393123626709,
|
| 648 |
+
"learning_rate": 3.770638107987506e-05,
|
| 649 |
+
"loss": 0.6561,
|
| 650 |
+
"step": 800
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 9.70059880239521,
|
| 654 |
+
"grad_norm": 5.898914337158203,
|
| 655 |
+
"learning_rate": 3.748326639892905e-05,
|
| 656 |
+
"loss": 0.6197,
|
| 657 |
+
"step": 810
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 9.820359281437126,
|
| 661 |
+
"grad_norm": 4.737303733825684,
|
| 662 |
+
"learning_rate": 3.7260151717983045e-05,
|
| 663 |
+
"loss": 0.6068,
|
| 664 |
+
"step": 820
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 9.940119760479043,
|
| 668 |
+
"grad_norm": 3.769287347793579,
|
| 669 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 670 |
+
"loss": 0.6261,
|
| 671 |
+
"step": 830
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 10.0,
|
| 675 |
+
"eval_accuracy": 0.7380239520958084,
|
| 676 |
+
"eval_loss": 0.568150520324707,
|
| 677 |
+
"eval_runtime": 5.0718,
|
| 678 |
+
"eval_samples_per_second": 131.708,
|
| 679 |
+
"eval_steps_per_second": 4.141,
|
| 680 |
+
"step": 835
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 10.059880239520957,
|
| 684 |
+
"grad_norm": 4.2985920906066895,
|
| 685 |
+
"learning_rate": 3.6813922356091035e-05,
|
| 686 |
+
"loss": 0.6455,
|
| 687 |
+
"step": 840
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 10.179640718562874,
|
| 691 |
+
"grad_norm": 4.344922065734863,
|
| 692 |
+
"learning_rate": 3.659080767514503e-05,
|
| 693 |
+
"loss": 0.6438,
|
| 694 |
+
"step": 850
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 10.29940119760479,
|
| 698 |
+
"grad_norm": 4.293480396270752,
|
| 699 |
+
"learning_rate": 3.636769299419902e-05,
|
| 700 |
+
"loss": 0.6161,
|
| 701 |
+
"step": 860
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 10.419161676646706,
|
| 705 |
+
"grad_norm": 5.124499797821045,
|
| 706 |
+
"learning_rate": 3.614457831325301e-05,
|
| 707 |
+
"loss": 0.5976,
|
| 708 |
+
"step": 870
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 10.538922155688622,
|
| 712 |
+
"grad_norm": 3.7405636310577393,
|
| 713 |
+
"learning_rate": 3.592146363230701e-05,
|
| 714 |
+
"loss": 0.6125,
|
| 715 |
+
"step": 880
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 10.658682634730539,
|
| 719 |
+
"grad_norm": 3.614593744277954,
|
| 720 |
+
"learning_rate": 3.5698348951361e-05,
|
| 721 |
+
"loss": 0.6159,
|
| 722 |
+
"step": 890
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 10.778443113772456,
|
| 726 |
+
"grad_norm": 3.6515111923217773,
|
| 727 |
+
"learning_rate": 3.5475234270415e-05,
|
| 728 |
+
"loss": 0.6095,
|
| 729 |
+
"step": 900
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 10.89820359281437,
|
| 733 |
+
"grad_norm": 4.4123215675354,
|
| 734 |
+
"learning_rate": 3.525211958946899e-05,
|
| 735 |
+
"loss": 0.6153,
|
| 736 |
+
"step": 910
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 10.994011976047904,
|
| 740 |
+
"eval_accuracy": 0.718562874251497,
|
| 741 |
+
"eval_loss": 0.6006675362586975,
|
| 742 |
+
"eval_runtime": 4.9154,
|
| 743 |
+
"eval_samples_per_second": 135.9,
|
| 744 |
+
"eval_steps_per_second": 4.272,
|
| 745 |
+
"step": 918
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 11.017964071856287,
|
| 749 |
+
"grad_norm": 3.9871232509613037,
|
| 750 |
+
"learning_rate": 3.502900490852298e-05,
|
| 751 |
+
"loss": 0.6124,
|
| 752 |
+
"step": 920
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 11.137724550898204,
|
| 756 |
+
"grad_norm": 4.505560874938965,
|
| 757 |
+
"learning_rate": 3.4805890227576974e-05,
|
| 758 |
+
"loss": 0.602,
|
| 759 |
+
"step": 930
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 11.25748502994012,
|
| 763 |
+
"grad_norm": 4.445052623748779,
|
| 764 |
+
"learning_rate": 3.4582775546630966e-05,
|
| 765 |
+
"loss": 0.6002,
|
| 766 |
+
"step": 940
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 11.377245508982035,
|
| 770 |
+
"grad_norm": 3.532015562057495,
|
| 771 |
+
"learning_rate": 3.4359660865684965e-05,
|
| 772 |
+
"loss": 0.6242,
|
| 773 |
+
"step": 950
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 11.497005988023952,
|
| 777 |
+
"grad_norm": 6.202284812927246,
|
| 778 |
+
"learning_rate": 3.413654618473896e-05,
|
| 779 |
+
"loss": 0.6113,
|
| 780 |
+
"step": 960
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 11.616766467065869,
|
| 784 |
+
"grad_norm": 5.8437089920043945,
|
| 785 |
+
"learning_rate": 3.3913431503792955e-05,
|
| 786 |
+
"loss": 0.6037,
|
| 787 |
+
"step": 970
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 11.736526946107784,
|
| 791 |
+
"grad_norm": 4.447215557098389,
|
| 792 |
+
"learning_rate": 3.369031682284695e-05,
|
| 793 |
+
"loss": 0.6356,
|
| 794 |
+
"step": 980
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 11.8562874251497,
|
| 798 |
+
"grad_norm": 4.645685195922852,
|
| 799 |
+
"learning_rate": 3.346720214190094e-05,
|
| 800 |
+
"loss": 0.6309,
|
| 801 |
+
"step": 990
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 11.976047904191617,
|
| 805 |
+
"grad_norm": 4.300328731536865,
|
| 806 |
+
"learning_rate": 3.324408746095493e-05,
|
| 807 |
+
"loss": 0.6046,
|
| 808 |
+
"step": 1000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 12.0,
|
| 812 |
+
"eval_accuracy": 0.7410179640718563,
|
| 813 |
+
"eval_loss": 0.5744568705558777,
|
| 814 |
+
"eval_runtime": 4.8113,
|
| 815 |
+
"eval_samples_per_second": 138.84,
|
| 816 |
+
"eval_steps_per_second": 4.365,
|
| 817 |
+
"step": 1002
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 12.095808383233534,
|
| 821 |
+
"grad_norm": 3.7963852882385254,
|
| 822 |
+
"learning_rate": 3.302097278000892e-05,
|
| 823 |
+
"loss": 0.5979,
|
| 824 |
+
"step": 1010
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 12.215568862275449,
|
| 828 |
+
"grad_norm": 5.136913776397705,
|
| 829 |
+
"learning_rate": 3.279785809906292e-05,
|
| 830 |
+
"loss": 0.6082,
|
| 831 |
+
"step": 1020
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 12.335329341317365,
|
| 835 |
+
"grad_norm": 5.207279205322266,
|
| 836 |
+
"learning_rate": 3.257474341811691e-05,
|
| 837 |
+
"loss": 0.5884,
|
| 838 |
+
"step": 1030
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 12.455089820359282,
|
| 842 |
+
"grad_norm": 4.387267589569092,
|
| 843 |
+
"learning_rate": 3.235162873717091e-05,
|
| 844 |
+
"loss": 0.6156,
|
| 845 |
+
"step": 1040
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 12.574850299401197,
|
| 849 |
+
"grad_norm": 3.0785038471221924,
|
| 850 |
+
"learning_rate": 3.21285140562249e-05,
|
| 851 |
+
"loss": 0.6093,
|
| 852 |
+
"step": 1050
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 12.694610778443113,
|
| 856 |
+
"grad_norm": 3.2741942405700684,
|
| 857 |
+
"learning_rate": 3.1905399375278894e-05,
|
| 858 |
+
"loss": 0.6033,
|
| 859 |
+
"step": 1060
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 12.81437125748503,
|
| 863 |
+
"grad_norm": 4.547226428985596,
|
| 864 |
+
"learning_rate": 3.1682284694332886e-05,
|
| 865 |
+
"loss": 0.6157,
|
| 866 |
+
"step": 1070
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 12.934131736526947,
|
| 870 |
+
"grad_norm": 4.921385765075684,
|
| 871 |
+
"learning_rate": 3.1459170013386885e-05,
|
| 872 |
+
"loss": 0.5679,
|
| 873 |
+
"step": 1080
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 12.994011976047904,
|
| 877 |
+
"eval_accuracy": 0.7230538922155688,
|
| 878 |
+
"eval_loss": 0.595708429813385,
|
| 879 |
+
"eval_runtime": 4.9064,
|
| 880 |
+
"eval_samples_per_second": 136.149,
|
| 881 |
+
"eval_steps_per_second": 4.28,
|
| 882 |
+
"step": 1085
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 13.053892215568862,
|
| 886 |
+
"grad_norm": 6.123495578765869,
|
| 887 |
+
"learning_rate": 3.1236055332440876e-05,
|
| 888 |
+
"loss": 0.5848,
|
| 889 |
+
"step": 1090
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 13.173652694610778,
|
| 893 |
+
"grad_norm": 3.932276964187622,
|
| 894 |
+
"learning_rate": 3.101294065149487e-05,
|
| 895 |
+
"loss": 0.6018,
|
| 896 |
+
"step": 1100
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 13.293413173652695,
|
| 900 |
+
"grad_norm": 4.827797889709473,
|
| 901 |
+
"learning_rate": 3.078982597054887e-05,
|
| 902 |
+
"loss": 0.6069,
|
| 903 |
+
"step": 1110
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 13.41317365269461,
|
| 907 |
+
"grad_norm": 3.6165199279785156,
|
| 908 |
+
"learning_rate": 3.056671128960286e-05,
|
| 909 |
+
"loss": 0.6036,
|
| 910 |
+
"step": 1120
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 13.532934131736527,
|
| 914 |
+
"grad_norm": 4.21213960647583,
|
| 915 |
+
"learning_rate": 3.034359660865685e-05,
|
| 916 |
+
"loss": 0.5872,
|
| 917 |
+
"step": 1130
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 13.652694610778443,
|
| 921 |
+
"grad_norm": 6.651663780212402,
|
| 922 |
+
"learning_rate": 3.012048192771085e-05,
|
| 923 |
+
"loss": 0.6057,
|
| 924 |
+
"step": 1140
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 13.77245508982036,
|
| 928 |
+
"grad_norm": 5.351555824279785,
|
| 929 |
+
"learning_rate": 2.989736724676484e-05,
|
| 930 |
+
"loss": 0.5951,
|
| 931 |
+
"step": 1150
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 13.892215568862275,
|
| 935 |
+
"grad_norm": 5.491767406463623,
|
| 936 |
+
"learning_rate": 2.9674252565818832e-05,
|
| 937 |
+
"loss": 0.6027,
|
| 938 |
+
"step": 1160
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 14.0,
|
| 942 |
+
"eval_accuracy": 0.7215568862275449,
|
| 943 |
+
"eval_loss": 0.5884155631065369,
|
| 944 |
+
"eval_runtime": 4.7316,
|
| 945 |
+
"eval_samples_per_second": 141.178,
|
| 946 |
+
"eval_steps_per_second": 4.438,
|
| 947 |
+
"step": 1169
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 14.011976047904191,
|
| 951 |
+
"grad_norm": 4.720607757568359,
|
| 952 |
+
"learning_rate": 2.9451137884872827e-05,
|
| 953 |
+
"loss": 0.5779,
|
| 954 |
+
"step": 1170
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 14.131736526946108,
|
| 958 |
+
"grad_norm": 4.377676963806152,
|
| 959 |
+
"learning_rate": 2.922802320392682e-05,
|
| 960 |
+
"loss": 0.5992,
|
| 961 |
+
"step": 1180
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 14.251497005988025,
|
| 965 |
+
"grad_norm": 4.529723644256592,
|
| 966 |
+
"learning_rate": 2.900490852298081e-05,
|
| 967 |
+
"loss": 0.6127,
|
| 968 |
+
"step": 1190
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 14.37125748502994,
|
| 972 |
+
"grad_norm": 3.385350465774536,
|
| 973 |
+
"learning_rate": 2.878179384203481e-05,
|
| 974 |
+
"loss": 0.5876,
|
| 975 |
+
"step": 1200
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 14.491017964071856,
|
| 979 |
+
"grad_norm": 5.14049768447876,
|
| 980 |
+
"learning_rate": 2.85586791610888e-05,
|
| 981 |
+
"loss": 0.5685,
|
| 982 |
+
"step": 1210
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 14.610778443113773,
|
| 986 |
+
"grad_norm": 6.226632118225098,
|
| 987 |
+
"learning_rate": 2.8335564480142796e-05,
|
| 988 |
+
"loss": 0.5957,
|
| 989 |
+
"step": 1220
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 14.730538922155688,
|
| 993 |
+
"grad_norm": 4.943429470062256,
|
| 994 |
+
"learning_rate": 2.8112449799196788e-05,
|
| 995 |
+
"loss": 0.5709,
|
| 996 |
+
"step": 1230
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 14.850299401197605,
|
| 1000 |
+
"grad_norm": 3.945502758026123,
|
| 1001 |
+
"learning_rate": 2.788933511825078e-05,
|
| 1002 |
+
"loss": 0.5807,
|
| 1003 |
+
"step": 1240
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 14.970059880239521,
|
| 1007 |
+
"grad_norm": 5.3703718185424805,
|
| 1008 |
+
"learning_rate": 2.7666220437304775e-05,
|
| 1009 |
+
"loss": 0.6249,
|
| 1010 |
+
"step": 1250
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 14.994011976047904,
|
| 1014 |
+
"eval_accuracy": 0.7365269461077845,
|
| 1015 |
+
"eval_loss": 0.5808472037315369,
|
| 1016 |
+
"eval_runtime": 4.8318,
|
| 1017 |
+
"eval_samples_per_second": 138.251,
|
| 1018 |
+
"eval_steps_per_second": 4.346,
|
| 1019 |
+
"step": 1252
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 15.089820359281438,
|
| 1023 |
+
"grad_norm": 3.6692726612091064,
|
| 1024 |
+
"learning_rate": 2.7443105756358774e-05,
|
| 1025 |
+
"loss": 0.5944,
|
| 1026 |
+
"step": 1260
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 15.209580838323353,
|
| 1030 |
+
"grad_norm": 4.581055164337158,
|
| 1031 |
+
"learning_rate": 2.7219991075412765e-05,
|
| 1032 |
+
"loss": 0.5599,
|
| 1033 |
+
"step": 1270
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 15.32934131736527,
|
| 1037 |
+
"grad_norm": 3.810741901397705,
|
| 1038 |
+
"learning_rate": 2.6996876394466757e-05,
|
| 1039 |
+
"loss": 0.577,
|
| 1040 |
+
"step": 1280
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 15.449101796407186,
|
| 1044 |
+
"grad_norm": 4.873391151428223,
|
| 1045 |
+
"learning_rate": 2.6773761713520752e-05,
|
| 1046 |
+
"loss": 0.565,
|
| 1047 |
+
"step": 1290
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 15.568862275449101,
|
| 1051 |
+
"grad_norm": 4.216405391693115,
|
| 1052 |
+
"learning_rate": 2.6550647032574744e-05,
|
| 1053 |
+
"loss": 0.5709,
|
| 1054 |
+
"step": 1300
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 15.688622754491018,
|
| 1058 |
+
"grad_norm": 4.651778697967529,
|
| 1059 |
+
"learning_rate": 2.6327532351628736e-05,
|
| 1060 |
+
"loss": 0.5984,
|
| 1061 |
+
"step": 1310
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 15.808383233532934,
|
| 1065 |
+
"grad_norm": 4.126067638397217,
|
| 1066 |
+
"learning_rate": 2.6104417670682734e-05,
|
| 1067 |
+
"loss": 0.5718,
|
| 1068 |
+
"step": 1320
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 15.928143712574851,
|
| 1072 |
+
"grad_norm": 3.843979597091675,
|
| 1073 |
+
"learning_rate": 2.5881302989736726e-05,
|
| 1074 |
+
"loss": 0.6059,
|
| 1075 |
+
"step": 1330
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 16.0,
|
| 1079 |
+
"eval_accuracy": 0.7350299401197605,
|
| 1080 |
+
"eval_loss": 0.5699232816696167,
|
| 1081 |
+
"eval_runtime": 5.1217,
|
| 1082 |
+
"eval_samples_per_second": 130.426,
|
| 1083 |
+
"eval_steps_per_second": 4.1,
|
| 1084 |
+
"step": 1336
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 16.047904191616766,
|
| 1088 |
+
"grad_norm": 3.6033108234405518,
|
| 1089 |
+
"learning_rate": 2.565818830879072e-05,
|
| 1090 |
+
"loss": 0.5773,
|
| 1091 |
+
"step": 1340
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 16.16766467065868,
|
| 1095 |
+
"grad_norm": 5.372034549713135,
|
| 1096 |
+
"learning_rate": 2.5435073627844713e-05,
|
| 1097 |
+
"loss": 0.5818,
|
| 1098 |
+
"step": 1350
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 16.2874251497006,
|
| 1102 |
+
"grad_norm": 4.0521955490112305,
|
| 1103 |
+
"learning_rate": 2.5211958946898705e-05,
|
| 1104 |
+
"loss": 0.5517,
|
| 1105 |
+
"step": 1360
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 16.407185628742514,
|
| 1109 |
+
"grad_norm": 4.455647945404053,
|
| 1110 |
+
"learning_rate": 2.4988844265952703e-05,
|
| 1111 |
+
"loss": 0.5904,
|
| 1112 |
+
"step": 1370
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 16.526946107784433,
|
| 1116 |
+
"grad_norm": 4.8622589111328125,
|
| 1117 |
+
"learning_rate": 2.4765729585006695e-05,
|
| 1118 |
+
"loss": 0.5943,
|
| 1119 |
+
"step": 1380
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 16.646706586826348,
|
| 1123 |
+
"grad_norm": 5.169972896575928,
|
| 1124 |
+
"learning_rate": 2.4542614904060687e-05,
|
| 1125 |
+
"loss": 0.5841,
|
| 1126 |
+
"step": 1390
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 16.766467065868262,
|
| 1130 |
+
"grad_norm": 4.88759708404541,
|
| 1131 |
+
"learning_rate": 2.4319500223114682e-05,
|
| 1132 |
+
"loss": 0.5639,
|
| 1133 |
+
"step": 1400
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 16.88622754491018,
|
| 1137 |
+
"grad_norm": 5.843952178955078,
|
| 1138 |
+
"learning_rate": 2.4096385542168677e-05,
|
| 1139 |
+
"loss": 0.5776,
|
| 1140 |
+
"step": 1410
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 16.994011976047904,
|
| 1144 |
+
"eval_accuracy": 0.7320359281437125,
|
| 1145 |
+
"eval_loss": 0.5769894123077393,
|
| 1146 |
+
"eval_runtime": 5.3536,
|
| 1147 |
+
"eval_samples_per_second": 124.775,
|
| 1148 |
+
"eval_steps_per_second": 3.923,
|
| 1149 |
+
"step": 1419
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 17.005988023952096,
|
| 1153 |
+
"grad_norm": 8.036992073059082,
|
| 1154 |
+
"learning_rate": 2.387327086122267e-05,
|
| 1155 |
+
"loss": 0.5775,
|
| 1156 |
+
"step": 1420
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 17.12574850299401,
|
| 1160 |
+
"grad_norm": 8.761232376098633,
|
| 1161 |
+
"learning_rate": 2.3650156180276664e-05,
|
| 1162 |
+
"loss": 0.56,
|
| 1163 |
+
"step": 1430
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 17.24550898203593,
|
| 1167 |
+
"grad_norm": 3.9877655506134033,
|
| 1168 |
+
"learning_rate": 2.3427041499330656e-05,
|
| 1169 |
+
"loss": 0.5409,
|
| 1170 |
+
"step": 1440
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 17.365269461077844,
|
| 1174 |
+
"grad_norm": 8.304950714111328,
|
| 1175 |
+
"learning_rate": 2.320392681838465e-05,
|
| 1176 |
+
"loss": 0.5696,
|
| 1177 |
+
"step": 1450
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 17.48502994011976,
|
| 1181 |
+
"grad_norm": 4.155704498291016,
|
| 1182 |
+
"learning_rate": 2.2980812137438646e-05,
|
| 1183 |
+
"loss": 0.5888,
|
| 1184 |
+
"step": 1460
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 17.604790419161677,
|
| 1188 |
+
"grad_norm": 4.174530029296875,
|
| 1189 |
+
"learning_rate": 2.2757697456492638e-05,
|
| 1190 |
+
"loss": 0.5566,
|
| 1191 |
+
"step": 1470
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 17.724550898203592,
|
| 1195 |
+
"grad_norm": 6.5803093910217285,
|
| 1196 |
+
"learning_rate": 2.253458277554663e-05,
|
| 1197 |
+
"loss": 0.5385,
|
| 1198 |
+
"step": 1480
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 17.84431137724551,
|
| 1202 |
+
"grad_norm": 4.418693542480469,
|
| 1203 |
+
"learning_rate": 2.2311468094600628e-05,
|
| 1204 |
+
"loss": 0.5411,
|
| 1205 |
+
"step": 1490
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 17.964071856287426,
|
| 1209 |
+
"grad_norm": 4.162361145019531,
|
| 1210 |
+
"learning_rate": 2.208835341365462e-05,
|
| 1211 |
+
"loss": 0.5903,
|
| 1212 |
+
"step": 1500
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 18.0,
|
| 1216 |
+
"eval_accuracy": 0.7215568862275449,
|
| 1217 |
+
"eval_loss": 0.5806027054786682,
|
| 1218 |
+
"eval_runtime": 4.6322,
|
| 1219 |
+
"eval_samples_per_second": 144.209,
|
| 1220 |
+
"eval_steps_per_second": 4.534,
|
| 1221 |
+
"step": 1503
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 18.08383233532934,
|
| 1225 |
+
"grad_norm": 4.274399280548096,
|
| 1226 |
+
"learning_rate": 2.186523873270861e-05,
|
| 1227 |
+
"loss": 0.5612,
|
| 1228 |
+
"step": 1510
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 18.20359281437126,
|
| 1232 |
+
"grad_norm": 5.013104438781738,
|
| 1233 |
+
"learning_rate": 2.1642124051762607e-05,
|
| 1234 |
+
"loss": 0.5654,
|
| 1235 |
+
"step": 1520
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 18.323353293413174,
|
| 1239 |
+
"grad_norm": 4.541128635406494,
|
| 1240 |
+
"learning_rate": 2.1419009370816602e-05,
|
| 1241 |
+
"loss": 0.5492,
|
| 1242 |
+
"step": 1530
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 18.44311377245509,
|
| 1246 |
+
"grad_norm": 4.409793853759766,
|
| 1247 |
+
"learning_rate": 2.1195894689870593e-05,
|
| 1248 |
+
"loss": 0.564,
|
| 1249 |
+
"step": 1540
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 18.562874251497007,
|
| 1253 |
+
"grad_norm": 3.6504085063934326,
|
| 1254 |
+
"learning_rate": 2.097278000892459e-05,
|
| 1255 |
+
"loss": 0.5531,
|
| 1256 |
+
"step": 1550
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 18.682634730538922,
|
| 1260 |
+
"grad_norm": 4.4471235275268555,
|
| 1261 |
+
"learning_rate": 2.074966532797858e-05,
|
| 1262 |
+
"loss": 0.5911,
|
| 1263 |
+
"step": 1560
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 18.802395209580837,
|
| 1267 |
+
"grad_norm": 4.53286600112915,
|
| 1268 |
+
"learning_rate": 2.0526550647032576e-05,
|
| 1269 |
+
"loss": 0.572,
|
| 1270 |
+
"step": 1570
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 18.922155688622755,
|
| 1274 |
+
"grad_norm": 4.156210899353027,
|
| 1275 |
+
"learning_rate": 2.030343596608657e-05,
|
| 1276 |
+
"loss": 0.5633,
|
| 1277 |
+
"step": 1580
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 18.994011976047904,
|
| 1281 |
+
"eval_accuracy": 0.7380239520958084,
|
| 1282 |
+
"eval_loss": 0.576755166053772,
|
| 1283 |
+
"eval_runtime": 5.1497,
|
| 1284 |
+
"eval_samples_per_second": 129.715,
|
| 1285 |
+
"eval_steps_per_second": 4.078,
|
| 1286 |
+
"step": 1586
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 19.04191616766467,
|
| 1290 |
+
"grad_norm": 4.36007022857666,
|
| 1291 |
+
"learning_rate": 2.0080321285140562e-05,
|
| 1292 |
+
"loss": 0.5304,
|
| 1293 |
+
"step": 1590
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 19.161676646706585,
|
| 1297 |
+
"grad_norm": 6.453935623168945,
|
| 1298 |
+
"learning_rate": 1.9857206604194558e-05,
|
| 1299 |
+
"loss": 0.5595,
|
| 1300 |
+
"step": 1600
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 19.281437125748504,
|
| 1304 |
+
"grad_norm": 4.35880708694458,
|
| 1305 |
+
"learning_rate": 1.9634091923248553e-05,
|
| 1306 |
+
"loss": 0.5578,
|
| 1307 |
+
"step": 1610
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 19.40119760479042,
|
| 1311 |
+
"grad_norm": 4.454617500305176,
|
| 1312 |
+
"learning_rate": 1.9410977242302544e-05,
|
| 1313 |
+
"loss": 0.532,
|
| 1314 |
+
"step": 1620
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 19.520958083832337,
|
| 1318 |
+
"grad_norm": 4.35628604888916,
|
| 1319 |
+
"learning_rate": 1.9187862561356536e-05,
|
| 1320 |
+
"loss": 0.5433,
|
| 1321 |
+
"step": 1630
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 19.64071856287425,
|
| 1325 |
+
"grad_norm": 4.488490104675293,
|
| 1326 |
+
"learning_rate": 1.896474788041053e-05,
|
| 1327 |
+
"loss": 0.5545,
|
| 1328 |
+
"step": 1640
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 19.760479041916167,
|
| 1332 |
+
"grad_norm": 5.309743881225586,
|
| 1333 |
+
"learning_rate": 1.8741633199464527e-05,
|
| 1334 |
+
"loss": 0.5793,
|
| 1335 |
+
"step": 1650
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 19.880239520958085,
|
| 1339 |
+
"grad_norm": 4.599639415740967,
|
| 1340 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 1341 |
+
"loss": 0.55,
|
| 1342 |
+
"step": 1660
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 20.0,
|
| 1346 |
+
"grad_norm": 4.7093892097473145,
|
| 1347 |
+
"learning_rate": 1.8295403837572513e-05,
|
| 1348 |
+
"loss": 0.5544,
|
| 1349 |
+
"step": 1670
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 20.0,
|
| 1353 |
+
"eval_accuracy": 0.7350299401197605,
|
| 1354 |
+
"eval_loss": 0.5829503536224365,
|
| 1355 |
+
"eval_runtime": 4.8934,
|
| 1356 |
+
"eval_samples_per_second": 136.511,
|
| 1357 |
+
"eval_steps_per_second": 4.292,
|
| 1358 |
+
"step": 1670
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 20.119760479041915,
|
| 1362 |
+
"grad_norm": 4.780275344848633,
|
| 1363 |
+
"learning_rate": 1.8072289156626505e-05,
|
| 1364 |
+
"loss": 0.5465,
|
| 1365 |
+
"step": 1680
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 20.239520958083833,
|
| 1369 |
+
"grad_norm": 4.53477668762207,
|
| 1370 |
+
"learning_rate": 1.78491744756805e-05,
|
| 1371 |
+
"loss": 0.525,
|
| 1372 |
+
"step": 1690
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 20.35928143712575,
|
| 1376 |
+
"grad_norm": 5.508608818054199,
|
| 1377 |
+
"learning_rate": 1.7626059794734495e-05,
|
| 1378 |
+
"loss": 0.5768,
|
| 1379 |
+
"step": 1700
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 20.479041916167663,
|
| 1383 |
+
"grad_norm": 5.174022197723389,
|
| 1384 |
+
"learning_rate": 1.7402945113788487e-05,
|
| 1385 |
+
"loss": 0.5308,
|
| 1386 |
+
"step": 1710
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 20.59880239520958,
|
| 1390 |
+
"grad_norm": 4.175419330596924,
|
| 1391 |
+
"learning_rate": 1.7179830432842482e-05,
|
| 1392 |
+
"loss": 0.5609,
|
| 1393 |
+
"step": 1720
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 20.718562874251496,
|
| 1397 |
+
"grad_norm": 4.944879055023193,
|
| 1398 |
+
"learning_rate": 1.6956715751896478e-05,
|
| 1399 |
+
"loss": 0.5323,
|
| 1400 |
+
"step": 1730
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 20.83832335329341,
|
| 1404 |
+
"grad_norm": 4.967810153961182,
|
| 1405 |
+
"learning_rate": 1.673360107095047e-05,
|
| 1406 |
+
"loss": 0.5234,
|
| 1407 |
+
"step": 1740
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 20.95808383233533,
|
| 1411 |
+
"grad_norm": 5.018576622009277,
|
| 1412 |
+
"learning_rate": 1.651048639000446e-05,
|
| 1413 |
+
"loss": 0.5515,
|
| 1414 |
+
"step": 1750
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 20.994011976047904,
|
| 1418 |
+
"eval_accuracy": 0.7260479041916168,
|
| 1419 |
+
"eval_loss": 0.5965989232063293,
|
| 1420 |
+
"eval_runtime": 4.6792,
|
| 1421 |
+
"eval_samples_per_second": 142.759,
|
| 1422 |
+
"eval_steps_per_second": 4.488,
|
| 1423 |
+
"step": 1753
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 21.077844311377245,
|
| 1427 |
+
"grad_norm": 4.204094886779785,
|
| 1428 |
+
"learning_rate": 1.6287371709058456e-05,
|
| 1429 |
+
"loss": 0.5672,
|
| 1430 |
+
"step": 1760
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 21.197604790419163,
|
| 1434 |
+
"grad_norm": 4.336850643157959,
|
| 1435 |
+
"learning_rate": 1.606425702811245e-05,
|
| 1436 |
+
"loss": 0.5011,
|
| 1437 |
+
"step": 1770
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 21.317365269461078,
|
| 1441 |
+
"grad_norm": 5.5377583503723145,
|
| 1442 |
+
"learning_rate": 1.5841142347166443e-05,
|
| 1443 |
+
"loss": 0.5755,
|
| 1444 |
+
"step": 1780
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 21.437125748502993,
|
| 1448 |
+
"grad_norm": 5.438154697418213,
|
| 1449 |
+
"learning_rate": 1.5618027666220438e-05,
|
| 1450 |
+
"loss": 0.5349,
|
| 1451 |
+
"step": 1790
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 21.55688622754491,
|
| 1455 |
+
"grad_norm": 5.496122360229492,
|
| 1456 |
+
"learning_rate": 1.5394912985274433e-05,
|
| 1457 |
+
"loss": 0.5754,
|
| 1458 |
+
"step": 1800
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 21.676646706586826,
|
| 1462 |
+
"grad_norm": 5.726380348205566,
|
| 1463 |
+
"learning_rate": 1.5171798304328425e-05,
|
| 1464 |
+
"loss": 0.5778,
|
| 1465 |
+
"step": 1810
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 21.79640718562874,
|
| 1469 |
+
"grad_norm": 4.841021537780762,
|
| 1470 |
+
"learning_rate": 1.494868362338242e-05,
|
| 1471 |
+
"loss": 0.5269,
|
| 1472 |
+
"step": 1820
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 21.91616766467066,
|
| 1476 |
+
"grad_norm": 4.8706583976745605,
|
| 1477 |
+
"learning_rate": 1.4725568942436414e-05,
|
| 1478 |
+
"loss": 0.5249,
|
| 1479 |
+
"step": 1830
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 22.0,
|
| 1483 |
+
"eval_accuracy": 0.7335329341317365,
|
| 1484 |
+
"eval_loss": 0.6078537106513977,
|
| 1485 |
+
"eval_runtime": 4.6418,
|
| 1486 |
+
"eval_samples_per_second": 143.909,
|
| 1487 |
+
"eval_steps_per_second": 4.524,
|
| 1488 |
+
"step": 1837
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 22.035928143712574,
|
| 1492 |
+
"grad_norm": 6.536875247955322,
|
| 1493 |
+
"learning_rate": 1.4502454261490405e-05,
|
| 1494 |
+
"loss": 0.5403,
|
| 1495 |
+
"step": 1840
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 22.15568862275449,
|
| 1499 |
+
"grad_norm": 5.156835079193115,
|
| 1500 |
+
"learning_rate": 1.42793395805444e-05,
|
| 1501 |
+
"loss": 0.5547,
|
| 1502 |
+
"step": 1850
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 22.275449101796408,
|
| 1506 |
+
"grad_norm": 4.475517749786377,
|
| 1507 |
+
"learning_rate": 1.4056224899598394e-05,
|
| 1508 |
+
"loss": 0.5474,
|
| 1509 |
+
"step": 1860
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 22.395209580838323,
|
| 1513 |
+
"grad_norm": 5.913077354431152,
|
| 1514 |
+
"learning_rate": 1.3833110218652387e-05,
|
| 1515 |
+
"loss": 0.5242,
|
| 1516 |
+
"step": 1870
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 22.51497005988024,
|
| 1520 |
+
"grad_norm": 4.135039806365967,
|
| 1521 |
+
"learning_rate": 1.3609995537706383e-05,
|
| 1522 |
+
"loss": 0.527,
|
| 1523 |
+
"step": 1880
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 22.634730538922156,
|
| 1527 |
+
"grad_norm": 4.773129940032959,
|
| 1528 |
+
"learning_rate": 1.3386880856760376e-05,
|
| 1529 |
+
"loss": 0.5828,
|
| 1530 |
+
"step": 1890
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 22.75449101796407,
|
| 1534 |
+
"grad_norm": 4.826302528381348,
|
| 1535 |
+
"learning_rate": 1.3163766175814368e-05,
|
| 1536 |
+
"loss": 0.5407,
|
| 1537 |
+
"step": 1900
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 22.87425149700599,
|
| 1541 |
+
"grad_norm": 5.572017669677734,
|
| 1542 |
+
"learning_rate": 1.2940651494868363e-05,
|
| 1543 |
+
"loss": 0.5329,
|
| 1544 |
+
"step": 1910
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 22.994011976047904,
|
| 1548 |
+
"grad_norm": 5.162022113800049,
|
| 1549 |
+
"learning_rate": 1.2717536813922356e-05,
|
| 1550 |
+
"loss": 0.5212,
|
| 1551 |
+
"step": 1920
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 22.994011976047904,
|
| 1555 |
+
"eval_accuracy": 0.7245508982035929,
|
| 1556 |
+
"eval_loss": 0.5972306132316589,
|
| 1557 |
+
"eval_runtime": 5.0266,
|
| 1558 |
+
"eval_samples_per_second": 132.892,
|
| 1559 |
+
"eval_steps_per_second": 4.178,
|
| 1560 |
+
"step": 1920
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 23.11377245508982,
|
| 1564 |
+
"grad_norm": 5.251833915710449,
|
| 1565 |
+
"learning_rate": 1.2494422132976352e-05,
|
| 1566 |
+
"loss": 0.5422,
|
| 1567 |
+
"step": 1930
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 23.233532934131738,
|
| 1571 |
+
"grad_norm": 4.400450706481934,
|
| 1572 |
+
"learning_rate": 1.2271307452030343e-05,
|
| 1573 |
+
"loss": 0.512,
|
| 1574 |
+
"step": 1940
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 23.353293413173652,
|
| 1578 |
+
"grad_norm": 5.726296901702881,
|
| 1579 |
+
"learning_rate": 1.2048192771084338e-05,
|
| 1580 |
+
"loss": 0.5167,
|
| 1581 |
+
"step": 1950
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 23.473053892215567,
|
| 1585 |
+
"grad_norm": 5.346691131591797,
|
| 1586 |
+
"learning_rate": 1.1825078090138332e-05,
|
| 1587 |
+
"loss": 0.5554,
|
| 1588 |
+
"step": 1960
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 23.592814371257486,
|
| 1592 |
+
"grad_norm": 6.358211994171143,
|
| 1593 |
+
"learning_rate": 1.1601963409192325e-05,
|
| 1594 |
+
"loss": 0.525,
|
| 1595 |
+
"step": 1970
|
| 1596 |
+
},
|
| 1597 |
+
{
|
| 1598 |
+
"epoch": 23.7125748502994,
|
| 1599 |
+
"grad_norm": 4.755873680114746,
|
| 1600 |
+
"learning_rate": 1.1378848728246319e-05,
|
| 1601 |
+
"loss": 0.5479,
|
| 1602 |
+
"step": 1980
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 23.83233532934132,
|
| 1606 |
+
"grad_norm": 6.708542346954346,
|
| 1607 |
+
"learning_rate": 1.1155734047300314e-05,
|
| 1608 |
+
"loss": 0.5457,
|
| 1609 |
+
"step": 1990
|
| 1610 |
+
},
|
| 1611 |
+
{
|
| 1612 |
+
"epoch": 23.952095808383234,
|
| 1613 |
+
"grad_norm": 4.143124103546143,
|
| 1614 |
+
"learning_rate": 1.0932619366354306e-05,
|
| 1615 |
+
"loss": 0.5268,
|
| 1616 |
+
"step": 2000
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 24.0,
|
| 1620 |
+
"eval_accuracy": 0.7230538922155688,
|
| 1621 |
+
"eval_loss": 0.5921865701675415,
|
| 1622 |
+
"eval_runtime": 5.3047,
|
| 1623 |
+
"eval_samples_per_second": 125.926,
|
| 1624 |
+
"eval_steps_per_second": 3.959,
|
| 1625 |
+
"step": 2004
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 24.07185628742515,
|
| 1629 |
+
"grad_norm": 5.585824489593506,
|
| 1630 |
+
"learning_rate": 1.0709504685408301e-05,
|
| 1631 |
+
"loss": 0.4723,
|
| 1632 |
+
"step": 2010
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 24.191616766467067,
|
| 1636 |
+
"grad_norm": 5.241277694702148,
|
| 1637 |
+
"learning_rate": 1.0486390004462294e-05,
|
| 1638 |
+
"loss": 0.5317,
|
| 1639 |
+
"step": 2020
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 24.311377245508982,
|
| 1643 |
+
"grad_norm": 5.840533256530762,
|
| 1644 |
+
"learning_rate": 1.0263275323516288e-05,
|
| 1645 |
+
"loss": 0.5458,
|
| 1646 |
+
"step": 2030
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 24.431137724550897,
|
| 1650 |
+
"grad_norm": 5.158961772918701,
|
| 1651 |
+
"learning_rate": 1.0040160642570281e-05,
|
| 1652 |
+
"loss": 0.5169,
|
| 1653 |
+
"step": 2040
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 24.550898203592816,
|
| 1657 |
+
"grad_norm": 4.132058620452881,
|
| 1658 |
+
"learning_rate": 9.817045961624276e-06,
|
| 1659 |
+
"loss": 0.531,
|
| 1660 |
+
"step": 2050
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 24.67065868263473,
|
| 1664 |
+
"grad_norm": 5.0692830085754395,
|
| 1665 |
+
"learning_rate": 9.593931280678268e-06,
|
| 1666 |
+
"loss": 0.5462,
|
| 1667 |
+
"step": 2060
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 24.790419161676645,
|
| 1671 |
+
"grad_norm": 5.38627815246582,
|
| 1672 |
+
"learning_rate": 9.370816599732263e-06,
|
| 1673 |
+
"loss": 0.5151,
|
| 1674 |
+
"step": 2070
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 24.910179640718564,
|
| 1678 |
+
"grad_norm": 5.605633735656738,
|
| 1679 |
+
"learning_rate": 9.147701918786257e-06,
|
| 1680 |
+
"loss": 0.5406,
|
| 1681 |
+
"step": 2080
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 24.994011976047904,
|
| 1685 |
+
"eval_accuracy": 0.7350299401197605,
|
| 1686 |
+
"eval_loss": 0.609959602355957,
|
| 1687 |
+
"eval_runtime": 5.5754,
|
| 1688 |
+
"eval_samples_per_second": 119.811,
|
| 1689 |
+
"eval_steps_per_second": 3.767,
|
| 1690 |
+
"step": 2087
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 25.02994011976048,
|
| 1694 |
+
"grad_norm": 5.846259117126465,
|
| 1695 |
+
"learning_rate": 8.92458723784025e-06,
|
| 1696 |
+
"loss": 0.5213,
|
| 1697 |
+
"step": 2090
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 25.149700598802394,
|
| 1701 |
+
"grad_norm": 7.1593708992004395,
|
| 1702 |
+
"learning_rate": 8.701472556894244e-06,
|
| 1703 |
+
"loss": 0.5329,
|
| 1704 |
+
"step": 2100
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 25.269461077844312,
|
| 1708 |
+
"grad_norm": 5.0290985107421875,
|
| 1709 |
+
"learning_rate": 8.478357875948239e-06,
|
| 1710 |
+
"loss": 0.5042,
|
| 1711 |
+
"step": 2110
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 25.389221556886227,
|
| 1715 |
+
"grad_norm": 4.644502639770508,
|
| 1716 |
+
"learning_rate": 8.25524319500223e-06,
|
| 1717 |
+
"loss": 0.5227,
|
| 1718 |
+
"step": 2120
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 25.508982035928145,
|
| 1722 |
+
"grad_norm": 5.292252540588379,
|
| 1723 |
+
"learning_rate": 8.032128514056226e-06,
|
| 1724 |
+
"loss": 0.5137,
|
| 1725 |
+
"step": 2130
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 25.62874251497006,
|
| 1729 |
+
"grad_norm": 4.673940658569336,
|
| 1730 |
+
"learning_rate": 7.809013833110219e-06,
|
| 1731 |
+
"loss": 0.5117,
|
| 1732 |
+
"step": 2140
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 25.748502994011975,
|
| 1736 |
+
"grad_norm": 5.361245632171631,
|
| 1737 |
+
"learning_rate": 7.5858991521642126e-06,
|
| 1738 |
+
"loss": 0.5067,
|
| 1739 |
+
"step": 2150
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 25.868263473053894,
|
| 1743 |
+
"grad_norm": 4.769536018371582,
|
| 1744 |
+
"learning_rate": 7.362784471218207e-06,
|
| 1745 |
+
"loss": 0.5562,
|
| 1746 |
+
"step": 2160
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 25.98802395209581,
|
| 1750 |
+
"grad_norm": 5.589327335357666,
|
| 1751 |
+
"learning_rate": 7.1396697902722e-06,
|
| 1752 |
+
"loss": 0.5257,
|
| 1753 |
+
"step": 2170
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 26.0,
|
| 1757 |
+
"eval_accuracy": 0.7305389221556886,
|
| 1758 |
+
"eval_loss": 0.6003913879394531,
|
| 1759 |
+
"eval_runtime": 5.6113,
|
| 1760 |
+
"eval_samples_per_second": 119.045,
|
| 1761 |
+
"eval_steps_per_second": 3.742,
|
| 1762 |
+
"step": 2171
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 26.107784431137723,
|
| 1766 |
+
"grad_norm": 5.76840353012085,
|
| 1767 |
+
"learning_rate": 6.916555109326194e-06,
|
| 1768 |
+
"loss": 0.5305,
|
| 1769 |
+
"step": 2180
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 26.227544910179642,
|
| 1773 |
+
"grad_norm": 4.287968158721924,
|
| 1774 |
+
"learning_rate": 6.693440428380188e-06,
|
| 1775 |
+
"loss": 0.5028,
|
| 1776 |
+
"step": 2190
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 26.347305389221557,
|
| 1780 |
+
"grad_norm": 7.93202543258667,
|
| 1781 |
+
"learning_rate": 6.4703257474341815e-06,
|
| 1782 |
+
"loss": 0.5173,
|
| 1783 |
+
"step": 2200
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"epoch": 26.46706586826347,
|
| 1787 |
+
"grad_norm": 5.515824794769287,
|
| 1788 |
+
"learning_rate": 6.247211066488176e-06,
|
| 1789 |
+
"loss": 0.5111,
|
| 1790 |
+
"step": 2210
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 26.58682634730539,
|
| 1794 |
+
"grad_norm": 7.206600666046143,
|
| 1795 |
+
"learning_rate": 6.024096385542169e-06,
|
| 1796 |
+
"loss": 0.5112,
|
| 1797 |
+
"step": 2220
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 26.706586826347305,
|
| 1801 |
+
"grad_norm": 4.152039527893066,
|
| 1802 |
+
"learning_rate": 5.800981704596163e-06,
|
| 1803 |
+
"loss": 0.4995,
|
| 1804 |
+
"step": 2230
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 26.82634730538922,
|
| 1808 |
+
"grad_norm": 5.948368072509766,
|
| 1809 |
+
"learning_rate": 5.577867023650157e-06,
|
| 1810 |
+
"loss": 0.5431,
|
| 1811 |
+
"step": 2240
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 26.94610778443114,
|
| 1815 |
+
"grad_norm": 4.189924240112305,
|
| 1816 |
+
"learning_rate": 5.3547523427041504e-06,
|
| 1817 |
+
"loss": 0.5152,
|
| 1818 |
+
"step": 2250
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 26.994011976047904,
|
| 1822 |
+
"eval_accuracy": 0.7320359281437125,
|
| 1823 |
+
"eval_loss": 0.6091659665107727,
|
| 1824 |
+
"eval_runtime": 4.8264,
|
| 1825 |
+
"eval_samples_per_second": 138.405,
|
| 1826 |
+
"eval_steps_per_second": 4.351,
|
| 1827 |
+
"step": 2254
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 27.065868263473053,
|
| 1831 |
+
"grad_norm": 5.771645545959473,
|
| 1832 |
+
"learning_rate": 5.131637661758144e-06,
|
| 1833 |
+
"loss": 0.5083,
|
| 1834 |
+
"step": 2260
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 27.18562874251497,
|
| 1838 |
+
"grad_norm": 5.393200397491455,
|
| 1839 |
+
"learning_rate": 4.908522980812138e-06,
|
| 1840 |
+
"loss": 0.4818,
|
| 1841 |
+
"step": 2270
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 27.305389221556887,
|
| 1845 |
+
"grad_norm": 5.9255828857421875,
|
| 1846 |
+
"learning_rate": 4.685408299866132e-06,
|
| 1847 |
+
"loss": 0.5215,
|
| 1848 |
+
"step": 2280
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 27.4251497005988,
|
| 1852 |
+
"grad_norm": 5.309273719787598,
|
| 1853 |
+
"learning_rate": 4.462293618920125e-06,
|
| 1854 |
+
"loss": 0.4627,
|
| 1855 |
+
"step": 2290
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 27.54491017964072,
|
| 1859 |
+
"grad_norm": 6.227340221405029,
|
| 1860 |
+
"learning_rate": 4.239178937974119e-06,
|
| 1861 |
+
"loss": 0.5067,
|
| 1862 |
+
"step": 2300
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 27.664670658682635,
|
| 1866 |
+
"grad_norm": 4.3603105545043945,
|
| 1867 |
+
"learning_rate": 4.016064257028113e-06,
|
| 1868 |
+
"loss": 0.5189,
|
| 1869 |
+
"step": 2310
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 27.78443113772455,
|
| 1873 |
+
"grad_norm": 5.04020357131958,
|
| 1874 |
+
"learning_rate": 3.7929495760821063e-06,
|
| 1875 |
+
"loss": 0.5311,
|
| 1876 |
+
"step": 2320
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 27.904191616766468,
|
| 1880 |
+
"grad_norm": 4.995678901672363,
|
| 1881 |
+
"learning_rate": 3.5698348951361e-06,
|
| 1882 |
+
"loss": 0.4858,
|
| 1883 |
+
"step": 2330
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 28.0,
|
| 1887 |
+
"eval_accuracy": 0.7230538922155688,
|
| 1888 |
+
"eval_loss": 0.6100460886955261,
|
| 1889 |
+
"eval_runtime": 5.2608,
|
| 1890 |
+
"eval_samples_per_second": 126.978,
|
| 1891 |
+
"eval_steps_per_second": 3.992,
|
| 1892 |
+
"step": 2338
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 28.023952095808383,
|
| 1896 |
+
"grad_norm": 3.8030707836151123,
|
| 1897 |
+
"learning_rate": 3.346720214190094e-06,
|
| 1898 |
+
"loss": 0.5329,
|
| 1899 |
+
"step": 2340
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 28.143712574850298,
|
| 1903 |
+
"grad_norm": 4.550446033477783,
|
| 1904 |
+
"learning_rate": 3.123605533244088e-06,
|
| 1905 |
+
"loss": 0.4682,
|
| 1906 |
+
"step": 2350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 28.263473053892216,
|
| 1910 |
+
"grad_norm": 4.607069492340088,
|
| 1911 |
+
"learning_rate": 2.9004908522980813e-06,
|
| 1912 |
+
"loss": 0.4928,
|
| 1913 |
+
"step": 2360
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 28.38323353293413,
|
| 1917 |
+
"grad_norm": 4.880204677581787,
|
| 1918 |
+
"learning_rate": 2.6773761713520752e-06,
|
| 1919 |
+
"loss": 0.5187,
|
| 1920 |
+
"step": 2370
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 28.50299401197605,
|
| 1924 |
+
"grad_norm": 4.848081111907959,
|
| 1925 |
+
"learning_rate": 2.454261490406069e-06,
|
| 1926 |
+
"loss": 0.5051,
|
| 1927 |
+
"step": 2380
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 28.622754491017965,
|
| 1931 |
+
"grad_norm": 5.9451189041137695,
|
| 1932 |
+
"learning_rate": 2.2311468094600625e-06,
|
| 1933 |
+
"loss": 0.5156,
|
| 1934 |
+
"step": 2390
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 28.74251497005988,
|
| 1938 |
+
"grad_norm": 5.852599143981934,
|
| 1939 |
+
"learning_rate": 2.0080321285140564e-06,
|
| 1940 |
+
"loss": 0.4913,
|
| 1941 |
+
"step": 2400
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 28.862275449101798,
|
| 1945 |
+
"grad_norm": 5.628122329711914,
|
| 1946 |
+
"learning_rate": 1.78491744756805e-06,
|
| 1947 |
+
"loss": 0.4711,
|
| 1948 |
+
"step": 2410
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 28.982035928143713,
|
| 1952 |
+
"grad_norm": 5.2325358390808105,
|
| 1953 |
+
"learning_rate": 1.561802766622044e-06,
|
| 1954 |
+
"loss": 0.5412,
|
| 1955 |
+
"step": 2420
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 28.994011976047904,
|
| 1959 |
+
"eval_accuracy": 0.7350299401197605,
|
| 1960 |
+
"eval_loss": 0.6115620136260986,
|
| 1961 |
+
"eval_runtime": 5.4339,
|
| 1962 |
+
"eval_samples_per_second": 122.932,
|
| 1963 |
+
"eval_steps_per_second": 3.865,
|
| 1964 |
+
"step": 2421
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 29.101796407185628,
|
| 1968 |
+
"grad_norm": 5.715780258178711,
|
| 1969 |
+
"learning_rate": 1.3386880856760376e-06,
|
| 1970 |
+
"loss": 0.528,
|
| 1971 |
+
"step": 2430
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 29.221556886227546,
|
| 1975 |
+
"grad_norm": 5.140947341918945,
|
| 1976 |
+
"learning_rate": 1.1155734047300313e-06,
|
| 1977 |
+
"loss": 0.5015,
|
| 1978 |
+
"step": 2440
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 29.34131736526946,
|
| 1982 |
+
"grad_norm": 5.15585470199585,
|
| 1983 |
+
"learning_rate": 8.92458723784025e-07,
|
| 1984 |
+
"loss": 0.4934,
|
| 1985 |
+
"step": 2450
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 29.461077844311376,
|
| 1989 |
+
"grad_norm": 3.845532178878784,
|
| 1990 |
+
"learning_rate": 6.693440428380188e-07,
|
| 1991 |
+
"loss": 0.4809,
|
| 1992 |
+
"step": 2460
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 29.580838323353294,
|
| 1996 |
+
"grad_norm": 5.481026649475098,
|
| 1997 |
+
"learning_rate": 4.462293618920125e-07,
|
| 1998 |
+
"loss": 0.4645,
|
| 1999 |
+
"step": 2470
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 29.70059880239521,
|
| 2003 |
+
"grad_norm": 4.100111484527588,
|
| 2004 |
+
"learning_rate": 2.2311468094600626e-07,
|
| 2005 |
+
"loss": 0.4856,
|
| 2006 |
+
"step": 2480
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 29.820359281437124,
|
| 2010 |
+
"grad_norm": 5.421600818634033,
|
| 2011 |
+
"learning_rate": 0.0,
|
| 2012 |
+
"loss": 0.4972,
|
| 2013 |
+
"step": 2490
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 29.820359281437124,
|
| 2017 |
+
"eval_accuracy": 0.7290419161676647,
|
| 2018 |
+
"eval_loss": 0.6119701862335205,
|
| 2019 |
+
"eval_runtime": 4.8411,
|
| 2020 |
+
"eval_samples_per_second": 137.985,
|
| 2021 |
+
"eval_steps_per_second": 4.338,
|
| 2022 |
+
"step": 2490
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 29.820359281437124,
|
| 2026 |
+
"step": 2490,
|
| 2027 |
+
"total_flos": 7.910788670992908e+18,
|
| 2028 |
+
"train_loss": 0.5937201630159554,
|
| 2029 |
+
"train_runtime": 3590.0203,
|
| 2030 |
+
"train_samples_per_second": 89.181,
|
| 2031 |
+
"train_steps_per_second": 0.694
|
| 2032 |
}
|
| 2033 |
],
|
| 2034 |
"logging_steps": 10,
|
| 2035 |
+
"max_steps": 2490,
|
| 2036 |
"num_input_tokens_seen": 0,
|
| 2037 |
+
"num_train_epochs": 30,
|
| 2038 |
"save_steps": 500,
|
| 2039 |
"stateful_callbacks": {
|
| 2040 |
"TrainerControl": {
|
|
|
|
| 2048 |
"attributes": {}
|
| 2049 |
}
|
| 2050 |
},
|
| 2051 |
+
"total_flos": 7.910788670992908e+18,
|
| 2052 |
"train_batch_size": 32,
|
| 2053 |
"trial_name": null,
|
| 2054 |
"trial_params": null
|