End of training
Browse files- README.md +4 -2
- all_results.json +8 -8
- eval_results.json +4 -4
- train_results.json +4 -4
- trainer_state.json +883 -157
README.md
CHANGED
|
@@ -3,6 +3,8 @@ license: mit
|
|
| 3 |
base_model: unicamp-dl/ptt5-base-t5-vocab
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
|
|
|
|
|
|
| 6 |
model-index:
|
| 7 |
- name: t5_base-qg-ap-nopeft
|
| 8 |
results: []
|
|
@@ -13,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# t5_base-qg-ap-nopeft
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [unicamp-dl/ptt5-base-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-base-t5-vocab) on
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 1.
|
| 19 |
|
| 20 |
## Model description
|
| 21 |
|
|
|
|
| 3 |
base_model: unicamp-dl/ptt5-base-t5-vocab
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
+
datasets:
|
| 7 |
+
- tiagoblima/qg_squad_v1_pt
|
| 8 |
model-index:
|
| 9 |
- name: t5_base-qg-ap-nopeft
|
| 10 |
results: []
|
|
|
|
| 15 |
|
| 16 |
# t5_base-qg-ap-nopeft
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [unicamp-dl/ptt5-base-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-base-t5-vocab) on the tiagoblima/qg_squad_v1_pt dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.1980
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_loss": 1.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 3585,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"train_loss": 1.
|
| 9 |
-
"train_runtime":
|
| 10 |
"train_samples": 51704,
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 1.1979700326919556,
|
| 4 |
+
"eval_runtime": 100.8452,
|
| 5 |
"eval_samples": 3585,
|
| 6 |
+
"eval_samples_per_second": 35.55,
|
| 7 |
+
"eval_steps_per_second": 8.895,
|
| 8 |
+
"train_loss": 1.0387100059157441,
|
| 9 |
+
"train_runtime": 30224.2981,
|
| 10 |
"train_samples": 51704,
|
| 11 |
+
"train_samples_per_second": 8.553,
|
| 12 |
+
"train_steps_per_second": 0.535
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"eval_loss": 1.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 3585,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"eval_loss": 1.1979700326919556,
|
| 4 |
+
"eval_runtime": 100.8452,
|
| 5 |
"eval_samples": 3585,
|
| 6 |
+
"eval_samples_per_second": 35.55,
|
| 7 |
+
"eval_steps_per_second": 8.895
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 51704,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"train_loss": 1.0387100059157441,
|
| 4 |
+
"train_runtime": 30224.2981,
|
| 5 |
"train_samples": 51704,
|
| 6 |
+
"train_samples_per_second": 8.553,
|
| 7 |
+
"train_steps_per_second": 0.535
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,308 +1,1034 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 1.
|
| 3 |
-
"best_model_checkpoint": "/temp/t5_base-qg-ap-nopeft/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"learning_rate": 9.
|
| 14 |
-
"loss":
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 9.
|
| 20 |
-
"loss": 1.
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 9.
|
| 26 |
-
"loss": 1.
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate": 9.
|
| 32 |
-
"loss": 1.
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss": 1.
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 1.
|
| 45 |
"step": 600
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 700
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss": 1.
|
| 57 |
"step": 800
|
| 58 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
{
|
| 60 |
"epoch": 1.0,
|
| 61 |
-
"eval_loss": 1.
|
| 62 |
-
"eval_runtime":
|
| 63 |
-
"eval_samples_per_second":
|
| 64 |
-
"eval_steps_per_second":
|
| 65 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 1.11,
|
| 69 |
-
"learning_rate": 7.
|
| 70 |
-
"loss": 1.
|
| 71 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
},
|
| 73 |
{
|
| 74 |
"epoch": 1.24,
|
| 75 |
-
"learning_rate": 7.
|
| 76 |
-
"loss": 1.
|
| 77 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 1.36,
|
| 81 |
-
"learning_rate": 7.
|
| 82 |
-
"loss": 1.
|
| 83 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 84 |
},
|
| 85 |
{
|
| 86 |
"epoch": 1.49,
|
| 87 |
-
"learning_rate": 7.
|
| 88 |
-
"loss": 1.
|
| 89 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 1.61,
|
| 93 |
-
"learning_rate": 6.
|
| 94 |
-
"loss": 1.
|
| 95 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 1.73,
|
| 99 |
-
"learning_rate": 6.
|
| 100 |
-
"loss": 1.
|
| 101 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.86,
|
| 105 |
-
"learning_rate": 6.
|
| 106 |
-
"loss": 1.
|
| 107 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 1.98,
|
| 111 |
-
"learning_rate": 6.
|
| 112 |
-
"loss": 1.
|
| 113 |
-
"step":
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 2.0,
|
| 117 |
-
"eval_loss": 1.
|
| 118 |
-
"eval_runtime":
|
| 119 |
-
"eval_samples_per_second":
|
| 120 |
-
"eval_steps_per_second":
|
| 121 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 2.1,
|
| 125 |
-
"learning_rate": 5.
|
| 126 |
-
"loss":
|
| 127 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
},
|
| 129 |
{
|
| 130 |
"epoch": 2.23,
|
| 131 |
-
"learning_rate": 5.
|
| 132 |
-
"loss":
|
| 133 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 134 |
},
|
| 135 |
{
|
| 136 |
"epoch": 2.35,
|
| 137 |
-
"learning_rate": 5.
|
| 138 |
-
"loss": 1.
|
| 139 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 140 |
},
|
| 141 |
{
|
| 142 |
"epoch": 2.48,
|
| 143 |
-
"learning_rate": 5.
|
| 144 |
-
"loss": 1.
|
| 145 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 2.6,
|
| 149 |
-
"learning_rate": 4.
|
| 150 |
-
"loss":
|
| 151 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 2.72,
|
| 155 |
-
"learning_rate": 4.
|
| 156 |
-
"loss":
|
| 157 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 2.85,
|
| 161 |
-
"learning_rate": 4.
|
| 162 |
-
"loss": 1.
|
| 163 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 2.97,
|
| 167 |
-
"learning_rate": 4.
|
| 168 |
-
"loss":
|
| 169 |
-
"step":
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 3.0,
|
| 173 |
-
"eval_loss": 1.
|
| 174 |
-
"eval_runtime":
|
| 175 |
-
"eval_samples_per_second":
|
| 176 |
-
"eval_steps_per_second":
|
| 177 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 3.09,
|
| 181 |
-
"learning_rate": 3.
|
| 182 |
-
"loss":
|
| 183 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 3.22,
|
| 187 |
-
"learning_rate": 3.
|
| 188 |
-
"loss":
|
| 189 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 3.34,
|
| 193 |
-
"learning_rate": 3.
|
| 194 |
-
"loss":
|
| 195 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 3.47,
|
| 199 |
-
"learning_rate": 3.
|
| 200 |
-
"loss":
|
| 201 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
},
|
| 203 |
{
|
| 204 |
"epoch": 3.59,
|
| 205 |
-
"learning_rate": 2.
|
| 206 |
-
"loss":
|
| 207 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 3.71,
|
| 211 |
-
"learning_rate": 2.
|
| 212 |
-
"loss":
|
| 213 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 3.84,
|
| 217 |
-
"learning_rate": 2.
|
| 218 |
-
"loss":
|
| 219 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 3.96,
|
| 223 |
-
"learning_rate": 2.
|
| 224 |
-
"loss":
|
| 225 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 4.0,
|
| 229 |
-
"eval_loss": 1.
|
| 230 |
-
"eval_runtime":
|
| 231 |
-
"eval_samples_per_second":
|
| 232 |
-
"eval_steps_per_second":
|
| 233 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 4.08,
|
| 237 |
-
"learning_rate": 1.
|
| 238 |
-
"loss":
|
| 239 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 4.21,
|
| 243 |
-
"learning_rate": 1.
|
| 244 |
-
"loss":
|
| 245 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 4.33,
|
| 249 |
-
"learning_rate": 1.
|
| 250 |
-
"loss":
|
| 251 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 4.46,
|
| 255 |
-
"learning_rate": 1.
|
| 256 |
-
"loss":
|
| 257 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 4.58,
|
| 261 |
-
"learning_rate": 8.
|
| 262 |
-
"loss":
|
| 263 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 264 |
},
|
| 265 |
{
|
| 266 |
"epoch": 4.7,
|
| 267 |
-
"learning_rate": 5.
|
| 268 |
-
"loss":
|
| 269 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 4.83,
|
| 273 |
-
"learning_rate": 3.
|
| 274 |
-
"loss":
|
| 275 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 4.95,
|
| 279 |
-
"learning_rate": 9.
|
| 280 |
-
"loss":
|
| 281 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 5.0,
|
| 285 |
-
"eval_loss": 1.
|
| 286 |
-
"eval_runtime":
|
| 287 |
-
"eval_samples_per_second":
|
| 288 |
-
"eval_steps_per_second":
|
| 289 |
-
"step":
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 5.0,
|
| 293 |
-
"step":
|
| 294 |
-
"total_flos": 1.
|
| 295 |
-
"train_loss": 1.
|
| 296 |
-
"train_runtime":
|
| 297 |
-
"train_samples_per_second":
|
| 298 |
-
"train_steps_per_second": 0.
|
| 299 |
}
|
| 300 |
],
|
| 301 |
"logging_steps": 100,
|
| 302 |
-
"max_steps":
|
| 303 |
"num_train_epochs": 5,
|
| 304 |
"save_steps": 500,
|
| 305 |
-
"total_flos": 1.
|
| 306 |
"trial_name": null,
|
| 307 |
"trial_params": null
|
| 308 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 1.1979700326919556,
|
| 3 |
+
"best_model_checkpoint": "/temp/t5_base-qg-ap-nopeft/checkpoint-9694",
|
| 4 |
+
"epoch": 4.9992263654649545,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 16155,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 9.938099659548128e-05,
|
| 14 |
+
"loss": 2.1468,
|
| 15 |
"step": 100
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.06,
|
| 19 |
+
"learning_rate": 9.876199319096255e-05,
|
| 20 |
+
"loss": 1.4194,
|
| 21 |
"step": 200
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.09,
|
| 25 |
+
"learning_rate": 9.814298978644383e-05,
|
| 26 |
+
"loss": 1.3613,
|
| 27 |
"step": 300
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.12,
|
| 31 |
+
"learning_rate": 9.752398638192511e-05,
|
| 32 |
+
"loss": 1.3139,
|
| 33 |
"step": 400
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.15,
|
| 37 |
+
"learning_rate": 9.690498297740637e-05,
|
| 38 |
+
"loss": 1.3216,
|
| 39 |
"step": 500
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.19,
|
| 43 |
+
"learning_rate": 9.628597957288765e-05,
|
| 44 |
+
"loss": 1.286,
|
| 45 |
"step": 600
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.22,
|
| 49 |
+
"learning_rate": 9.566697616836893e-05,
|
| 50 |
+
"loss": 1.2877,
|
| 51 |
"step": 700
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.25,
|
| 55 |
+
"learning_rate": 9.50479727638502e-05,
|
| 56 |
+
"loss": 1.246,
|
| 57 |
"step": 800
|
| 58 |
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.28,
|
| 61 |
+
"learning_rate": 9.442896935933148e-05,
|
| 62 |
+
"loss": 1.2309,
|
| 63 |
+
"step": 900
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.31,
|
| 67 |
+
"learning_rate": 9.380996595481276e-05,
|
| 68 |
+
"loss": 1.2611,
|
| 69 |
+
"step": 1000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.34,
|
| 73 |
+
"learning_rate": 9.319096255029403e-05,
|
| 74 |
+
"loss": 1.247,
|
| 75 |
+
"step": 1100
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.37,
|
| 79 |
+
"learning_rate": 9.257195914577531e-05,
|
| 80 |
+
"loss": 1.2398,
|
| 81 |
+
"step": 1200
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.4,
|
| 85 |
+
"learning_rate": 9.195295574125658e-05,
|
| 86 |
+
"loss": 1.2403,
|
| 87 |
+
"step": 1300
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.43,
|
| 91 |
+
"learning_rate": 9.133395233673786e-05,
|
| 92 |
+
"loss": 1.2324,
|
| 93 |
+
"step": 1400
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.46,
|
| 97 |
+
"learning_rate": 9.071494893221914e-05,
|
| 98 |
+
"loss": 1.2069,
|
| 99 |
+
"step": 1500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.5,
|
| 103 |
+
"learning_rate": 9.00959455277004e-05,
|
| 104 |
+
"loss": 1.2273,
|
| 105 |
+
"step": 1600
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.53,
|
| 109 |
+
"learning_rate": 8.947694212318168e-05,
|
| 110 |
+
"loss": 1.2199,
|
| 111 |
+
"step": 1700
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.56,
|
| 115 |
+
"learning_rate": 8.885793871866296e-05,
|
| 116 |
+
"loss": 1.1963,
|
| 117 |
+
"step": 1800
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.59,
|
| 121 |
+
"learning_rate": 8.823893531414423e-05,
|
| 122 |
+
"loss": 1.2081,
|
| 123 |
+
"step": 1900
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.62,
|
| 127 |
+
"learning_rate": 8.76199319096255e-05,
|
| 128 |
+
"loss": 1.1895,
|
| 129 |
+
"step": 2000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.65,
|
| 133 |
+
"learning_rate": 8.700092850510679e-05,
|
| 134 |
+
"loss": 1.203,
|
| 135 |
+
"step": 2100
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.68,
|
| 139 |
+
"learning_rate": 8.638192510058805e-05,
|
| 140 |
+
"loss": 1.1738,
|
| 141 |
+
"step": 2200
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.71,
|
| 145 |
+
"learning_rate": 8.576292169606933e-05,
|
| 146 |
+
"loss": 1.187,
|
| 147 |
+
"step": 2300
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.74,
|
| 151 |
+
"learning_rate": 8.514391829155061e-05,
|
| 152 |
+
"loss": 1.1857,
|
| 153 |
+
"step": 2400
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.77,
|
| 157 |
+
"learning_rate": 8.452491488703188e-05,
|
| 158 |
+
"loss": 1.169,
|
| 159 |
+
"step": 2500
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.8,
|
| 163 |
+
"learning_rate": 8.390591148251315e-05,
|
| 164 |
+
"loss": 1.2174,
|
| 165 |
+
"step": 2600
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.84,
|
| 169 |
+
"learning_rate": 8.328690807799443e-05,
|
| 170 |
+
"loss": 1.1737,
|
| 171 |
+
"step": 2700
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.87,
|
| 175 |
+
"learning_rate": 8.26679046734757e-05,
|
| 176 |
+
"loss": 1.174,
|
| 177 |
+
"step": 2800
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.9,
|
| 181 |
+
"learning_rate": 8.204890126895698e-05,
|
| 182 |
+
"loss": 1.1709,
|
| 183 |
+
"step": 2900
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.93,
|
| 187 |
+
"learning_rate": 8.142989786443826e-05,
|
| 188 |
+
"loss": 1.1882,
|
| 189 |
+
"step": 3000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.96,
|
| 193 |
+
"learning_rate": 8.081089445991952e-05,
|
| 194 |
+
"loss": 1.1793,
|
| 195 |
+
"step": 3100
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.99,
|
| 199 |
+
"learning_rate": 8.019189105540082e-05,
|
| 200 |
+
"loss": 1.1634,
|
| 201 |
+
"step": 3200
|
| 202 |
+
},
|
| 203 |
{
|
| 204 |
"epoch": 1.0,
|
| 205 |
+
"eval_loss": 1.224627137184143,
|
| 206 |
+
"eval_runtime": 100.7601,
|
| 207 |
+
"eval_samples_per_second": 35.58,
|
| 208 |
+
"eval_steps_per_second": 8.902,
|
| 209 |
+
"step": 3231
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.02,
|
| 213 |
+
"learning_rate": 7.95728876508821e-05,
|
| 214 |
+
"loss": 1.101,
|
| 215 |
+
"step": 3300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 1.05,
|
| 219 |
+
"learning_rate": 7.895388424636336e-05,
|
| 220 |
+
"loss": 1.0723,
|
| 221 |
+
"step": 3400
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 1.08,
|
| 225 |
+
"learning_rate": 7.833488084184464e-05,
|
| 226 |
+
"loss": 1.0846,
|
| 227 |
+
"step": 3500
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 1.11,
|
| 231 |
+
"learning_rate": 7.771587743732592e-05,
|
| 232 |
+
"loss": 1.0802,
|
| 233 |
+
"step": 3600
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.14,
|
| 237 |
+
"learning_rate": 7.709687403280718e-05,
|
| 238 |
+
"loss": 1.0806,
|
| 239 |
+
"step": 3700
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.18,
|
| 243 |
+
"learning_rate": 7.647787062828846e-05,
|
| 244 |
+
"loss": 1.0844,
|
| 245 |
+
"step": 3800
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.21,
|
| 249 |
+
"learning_rate": 7.585886722376974e-05,
|
| 250 |
+
"loss": 1.0838,
|
| 251 |
+
"step": 3900
|
| 252 |
},
|
| 253 |
{
|
| 254 |
"epoch": 1.24,
|
| 255 |
+
"learning_rate": 7.523986381925101e-05,
|
| 256 |
+
"loss": 1.0977,
|
| 257 |
+
"step": 4000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.27,
|
| 261 |
+
"learning_rate": 7.462086041473229e-05,
|
| 262 |
+
"loss": 1.0832,
|
| 263 |
+
"step": 4100
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.3,
|
| 267 |
+
"learning_rate": 7.400185701021357e-05,
|
| 268 |
+
"loss": 1.074,
|
| 269 |
+
"step": 4200
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.33,
|
| 273 |
+
"learning_rate": 7.338285360569483e-05,
|
| 274 |
+
"loss": 1.0835,
|
| 275 |
+
"step": 4300
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 1.36,
|
| 279 |
+
"learning_rate": 7.276385020117611e-05,
|
| 280 |
+
"loss": 1.0725,
|
| 281 |
+
"step": 4400
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.39,
|
| 285 |
+
"learning_rate": 7.214484679665738e-05,
|
| 286 |
+
"loss": 1.0854,
|
| 287 |
+
"step": 4500
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.42,
|
| 291 |
+
"learning_rate": 7.152584339213866e-05,
|
| 292 |
+
"loss": 1.0879,
|
| 293 |
+
"step": 4600
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.45,
|
| 297 |
+
"learning_rate": 7.090683998761994e-05,
|
| 298 |
+
"loss": 1.0916,
|
| 299 |
+
"step": 4700
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 1.49,
|
| 303 |
+
"learning_rate": 7.02878365831012e-05,
|
| 304 |
+
"loss": 1.0758,
|
| 305 |
+
"step": 4800
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.52,
|
| 309 |
+
"learning_rate": 6.966883317858248e-05,
|
| 310 |
+
"loss": 1.0953,
|
| 311 |
+
"step": 4900
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.55,
|
| 315 |
+
"learning_rate": 6.904982977406376e-05,
|
| 316 |
+
"loss": 1.0733,
|
| 317 |
+
"step": 5000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.58,
|
| 321 |
+
"learning_rate": 6.843082636954502e-05,
|
| 322 |
+
"loss": 1.0782,
|
| 323 |
+
"step": 5100
|
| 324 |
},
|
| 325 |
{
|
| 326 |
"epoch": 1.61,
|
| 327 |
+
"learning_rate": 6.78118229650263e-05,
|
| 328 |
+
"loss": 1.0785,
|
| 329 |
+
"step": 5200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.64,
|
| 333 |
+
"learning_rate": 6.719281956050758e-05,
|
| 334 |
+
"loss": 1.0708,
|
| 335 |
+
"step": 5300
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.67,
|
| 339 |
+
"learning_rate": 6.657381615598886e-05,
|
| 340 |
+
"loss": 1.0541,
|
| 341 |
+
"step": 5400
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 1.7,
|
| 345 |
+
"learning_rate": 6.595481275147014e-05,
|
| 346 |
+
"loss": 1.0629,
|
| 347 |
+
"step": 5500
|
| 348 |
},
|
| 349 |
{
|
| 350 |
"epoch": 1.73,
|
| 351 |
+
"learning_rate": 6.533580934695142e-05,
|
| 352 |
+
"loss": 1.0707,
|
| 353 |
+
"step": 5600
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.76,
|
| 357 |
+
"learning_rate": 6.471680594243269e-05,
|
| 358 |
+
"loss": 1.077,
|
| 359 |
+
"step": 5700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.79,
|
| 363 |
+
"learning_rate": 6.409780253791397e-05,
|
| 364 |
+
"loss": 1.0926,
|
| 365 |
+
"step": 5800
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 1.83,
|
| 369 |
+
"learning_rate": 6.347879913339524e-05,
|
| 370 |
+
"loss": 1.062,
|
| 371 |
+
"step": 5900
|
| 372 |
},
|
| 373 |
{
|
| 374 |
"epoch": 1.86,
|
| 375 |
+
"learning_rate": 6.285979572887651e-05,
|
| 376 |
+
"loss": 1.079,
|
| 377 |
+
"step": 6000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.89,
|
| 381 |
+
"learning_rate": 6.224079232435779e-05,
|
| 382 |
+
"loss": 1.065,
|
| 383 |
+
"step": 6100
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.92,
|
| 387 |
+
"learning_rate": 6.162178891983907e-05,
|
| 388 |
+
"loss": 1.0766,
|
| 389 |
+
"step": 6200
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.95,
|
| 393 |
+
"learning_rate": 6.100278551532034e-05,
|
| 394 |
+
"loss": 1.0814,
|
| 395 |
+
"step": 6300
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 1.98,
|
| 399 |
+
"learning_rate": 6.038378211080161e-05,
|
| 400 |
+
"loss": 1.0645,
|
| 401 |
+
"step": 6400
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 2.0,
|
| 405 |
+
"eval_loss": 1.2034717798233032,
|
| 406 |
+
"eval_runtime": 100.7994,
|
| 407 |
+
"eval_samples_per_second": 35.566,
|
| 408 |
+
"eval_steps_per_second": 8.899,
|
| 409 |
+
"step": 6463
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 2.01,
|
| 413 |
+
"learning_rate": 5.9764778706282886e-05,
|
| 414 |
+
"loss": 1.0357,
|
| 415 |
+
"step": 6500
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 2.04,
|
| 419 |
+
"learning_rate": 5.914577530176416e-05,
|
| 420 |
+
"loss": 0.9951,
|
| 421 |
+
"step": 6600
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.07,
|
| 425 |
+
"learning_rate": 5.852677189724544e-05,
|
| 426 |
+
"loss": 0.9866,
|
| 427 |
+
"step": 6700
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 2.1,
|
| 431 |
+
"learning_rate": 5.790776849272671e-05,
|
| 432 |
+
"loss": 0.9925,
|
| 433 |
+
"step": 6800
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 2.14,
|
| 437 |
+
"learning_rate": 5.728876508820798e-05,
|
| 438 |
+
"loss": 1.006,
|
| 439 |
+
"step": 6900
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.17,
|
| 443 |
+
"learning_rate": 5.666976168368926e-05,
|
| 444 |
+
"loss": 1.0164,
|
| 445 |
+
"step": 7000
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.2,
|
| 449 |
+
"learning_rate": 5.6050758279170534e-05,
|
| 450 |
+
"loss": 1.0098,
|
| 451 |
+
"step": 7100
|
| 452 |
},
|
| 453 |
{
|
| 454 |
"epoch": 2.23,
|
| 455 |
+
"learning_rate": 5.5431754874651806e-05,
|
| 456 |
+
"loss": 0.9917,
|
| 457 |
+
"step": 7200
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 2.26,
|
| 461 |
+
"learning_rate": 5.4812751470133085e-05,
|
| 462 |
+
"loss": 0.9891,
|
| 463 |
+
"step": 7300
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 2.29,
|
| 467 |
+
"learning_rate": 5.419374806561436e-05,
|
| 468 |
+
"loss": 0.9932,
|
| 469 |
+
"step": 7400
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 2.32,
|
| 473 |
+
"learning_rate": 5.357474466109564e-05,
|
| 474 |
+
"loss": 0.9881,
|
| 475 |
+
"step": 7500
|
| 476 |
},
|
| 477 |
{
|
| 478 |
"epoch": 2.35,
|
| 479 |
+
"learning_rate": 5.2955741256576916e-05,
|
| 480 |
+
"loss": 1.0001,
|
| 481 |
+
"step": 7600
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 2.38,
|
| 485 |
+
"learning_rate": 5.2336737852058195e-05,
|
| 486 |
+
"loss": 0.9836,
|
| 487 |
+
"step": 7700
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.41,
|
| 491 |
+
"learning_rate": 5.171773444753947e-05,
|
| 492 |
+
"loss": 0.9965,
|
| 493 |
+
"step": 7800
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 2.44,
|
| 497 |
+
"learning_rate": 5.109873104302074e-05,
|
| 498 |
+
"loss": 1.0102,
|
| 499 |
+
"step": 7900
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 2.48,
|
| 503 |
+
"learning_rate": 5.047972763850202e-05,
|
| 504 |
+
"loss": 1.0023,
|
| 505 |
+
"step": 8000
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 2.51,
|
| 509 |
+
"learning_rate": 4.986072423398329e-05,
|
| 510 |
+
"loss": 1.0118,
|
| 511 |
+
"step": 8100
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 2.54,
|
| 515 |
+
"learning_rate": 4.9241720829464564e-05,
|
| 516 |
+
"loss": 0.9785,
|
| 517 |
+
"step": 8200
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 2.57,
|
| 521 |
+
"learning_rate": 4.862271742494584e-05,
|
| 522 |
+
"loss": 1.0105,
|
| 523 |
+
"step": 8300
|
| 524 |
},
|
| 525 |
{
|
| 526 |
"epoch": 2.6,
|
| 527 |
+
"learning_rate": 4.8003714020427115e-05,
|
| 528 |
+
"loss": 0.998,
|
| 529 |
+
"step": 8400
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.63,
|
| 533 |
+
"learning_rate": 4.738471061590839e-05,
|
| 534 |
+
"loss": 0.9986,
|
| 535 |
+
"step": 8500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 2.66,
|
| 539 |
+
"learning_rate": 4.676570721138967e-05,
|
| 540 |
+
"loss": 0.9938,
|
| 541 |
+
"step": 8600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.69,
|
| 545 |
+
"learning_rate": 4.614670380687094e-05,
|
| 546 |
+
"loss": 0.9939,
|
| 547 |
+
"step": 8700
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 2.72,
|
| 551 |
+
"learning_rate": 4.552770040235221e-05,
|
| 552 |
+
"loss": 0.9846,
|
| 553 |
+
"step": 8800
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 2.75,
|
| 557 |
+
"learning_rate": 4.490869699783349e-05,
|
| 558 |
+
"loss": 1.0124,
|
| 559 |
+
"step": 8900
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 2.79,
|
| 563 |
+
"learning_rate": 4.428969359331476e-05,
|
| 564 |
+
"loss": 0.9964,
|
| 565 |
+
"step": 9000
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 2.82,
|
| 569 |
+
"learning_rate": 4.367069018879604e-05,
|
| 570 |
+
"loss": 1.0029,
|
| 571 |
+
"step": 9100
|
| 572 |
},
|
| 573 |
{
|
| 574 |
"epoch": 2.85,
|
| 575 |
+
"learning_rate": 4.3051686784277315e-05,
|
| 576 |
+
"loss": 1.0148,
|
| 577 |
+
"step": 9200
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 2.88,
|
| 581 |
+
"learning_rate": 4.2432683379758594e-05,
|
| 582 |
+
"loss": 1.0039,
|
| 583 |
+
"step": 9300
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 2.91,
|
| 587 |
+
"learning_rate": 4.1813679975239866e-05,
|
| 588 |
+
"loss": 0.9917,
|
| 589 |
+
"step": 9400
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 2.94,
|
| 593 |
+
"learning_rate": 4.119467657072114e-05,
|
| 594 |
+
"loss": 1.0066,
|
| 595 |
+
"step": 9500
|
| 596 |
},
|
| 597 |
{
|
| 598 |
"epoch": 2.97,
|
| 599 |
+
"learning_rate": 4.057567316620242e-05,
|
| 600 |
+
"loss": 0.991,
|
| 601 |
+
"step": 9600
|
| 602 |
},
|
| 603 |
{
|
| 604 |
"epoch": 3.0,
|
| 605 |
+
"eval_loss": 1.1979700326919556,
|
| 606 |
+
"eval_runtime": 100.8034,
|
| 607 |
+
"eval_samples_per_second": 35.564,
|
| 608 |
+
"eval_steps_per_second": 8.899,
|
| 609 |
+
"step": 9694
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 3.0,
|
| 613 |
+
"learning_rate": 3.995666976168369e-05,
|
| 614 |
+
"loss": 0.9994,
|
| 615 |
+
"step": 9700
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 3.03,
|
| 619 |
+
"learning_rate": 3.933766635716496e-05,
|
| 620 |
+
"loss": 0.9555,
|
| 621 |
+
"step": 9800
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 3.06,
|
| 625 |
+
"learning_rate": 3.871866295264624e-05,
|
| 626 |
+
"loss": 0.9364,
|
| 627 |
+
"step": 9900
|
| 628 |
},
|
| 629 |
{
|
| 630 |
"epoch": 3.09,
|
| 631 |
+
"learning_rate": 3.8099659548127514e-05,
|
| 632 |
+
"loss": 0.9392,
|
| 633 |
+
"step": 10000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 3.13,
|
| 637 |
+
"learning_rate": 3.748065614360879e-05,
|
| 638 |
+
"loss": 0.9439,
|
| 639 |
+
"step": 10100
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 3.16,
|
| 643 |
+
"learning_rate": 3.686165273909007e-05,
|
| 644 |
+
"loss": 0.9474,
|
| 645 |
+
"step": 10200
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 3.19,
|
| 649 |
+
"learning_rate": 3.6242649334571345e-05,
|
| 650 |
+
"loss": 0.9298,
|
| 651 |
+
"step": 10300
|
| 652 |
},
|
| 653 |
{
|
| 654 |
"epoch": 3.22,
|
| 655 |
+
"learning_rate": 3.562364593005262e-05,
|
| 656 |
+
"loss": 0.9245,
|
| 657 |
+
"step": 10400
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 3.25,
|
| 661 |
+
"learning_rate": 3.5004642525533896e-05,
|
| 662 |
+
"loss": 0.9724,
|
| 663 |
+
"step": 10500
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 3.28,
|
| 667 |
+
"learning_rate": 3.438563912101517e-05,
|
| 668 |
+
"loss": 0.9484,
|
| 669 |
+
"step": 10600
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 3.31,
|
| 673 |
+
"learning_rate": 3.376663571649644e-05,
|
| 674 |
+
"loss": 0.9376,
|
| 675 |
+
"step": 10700
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 3.34,
|
| 679 |
+
"learning_rate": 3.314763231197771e-05,
|
| 680 |
+
"loss": 0.9549,
|
| 681 |
+
"step": 10800
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 3.37,
|
| 685 |
+
"learning_rate": 3.252862890745899e-05,
|
| 686 |
+
"loss": 0.957,
|
| 687 |
+
"step": 10900
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 3.4,
|
| 691 |
+
"learning_rate": 3.1909625502940265e-05,
|
| 692 |
+
"loss": 0.9509,
|
| 693 |
+
"step": 11000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 3.43,
|
| 697 |
+
"learning_rate": 3.129062209842154e-05,
|
| 698 |
+
"loss": 0.9427,
|
| 699 |
+
"step": 11100
|
| 700 |
},
|
| 701 |
{
|
| 702 |
"epoch": 3.47,
|
| 703 |
+
"learning_rate": 3.067161869390282e-05,
|
| 704 |
+
"loss": 0.9598,
|
| 705 |
+
"step": 11200
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.5,
|
| 709 |
+
"learning_rate": 3.0052615289384095e-05,
|
| 710 |
+
"loss": 0.9584,
|
| 711 |
+
"step": 11300
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.53,
|
| 715 |
+
"learning_rate": 2.943361188486537e-05,
|
| 716 |
+
"loss": 0.9544,
|
| 717 |
+
"step": 11400
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 3.56,
|
| 721 |
+
"learning_rate": 2.8814608480346644e-05,
|
| 722 |
+
"loss": 0.9561,
|
| 723 |
+
"step": 11500
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 3.59,
|
| 727 |
+
"learning_rate": 2.819560507582792e-05,
|
| 728 |
+
"loss": 0.9374,
|
| 729 |
+
"step": 11600
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 3.62,
|
| 733 |
+
"learning_rate": 2.7576601671309192e-05,
|
| 734 |
+
"loss": 0.9236,
|
| 735 |
+
"step": 11700
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 3.65,
|
| 739 |
+
"learning_rate": 2.6957598266790468e-05,
|
| 740 |
+
"loss": 0.9434,
|
| 741 |
+
"step": 11800
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 3.68,
|
| 745 |
+
"learning_rate": 2.6338594862271743e-05,
|
| 746 |
+
"loss": 0.9363,
|
| 747 |
+
"step": 11900
|
| 748 |
},
|
| 749 |
{
|
| 750 |
"epoch": 3.71,
|
| 751 |
+
"learning_rate": 2.5719591457753016e-05,
|
| 752 |
+
"loss": 0.9381,
|
| 753 |
+
"step": 12000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 3.74,
|
| 757 |
+
"learning_rate": 2.510058805323429e-05,
|
| 758 |
+
"loss": 0.9535,
|
| 759 |
+
"step": 12100
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 3.78,
|
| 763 |
+
"learning_rate": 2.448158464871557e-05,
|
| 764 |
+
"loss": 0.9577,
|
| 765 |
+
"step": 12200
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 3.81,
|
| 769 |
+
"learning_rate": 2.3862581244196843e-05,
|
| 770 |
+
"loss": 0.9373,
|
| 771 |
+
"step": 12300
|
| 772 |
},
|
| 773 |
{
|
| 774 |
"epoch": 3.84,
|
| 775 |
+
"learning_rate": 2.324357783967812e-05,
|
| 776 |
+
"loss": 0.9499,
|
| 777 |
+
"step": 12400
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 3.87,
|
| 781 |
+
"learning_rate": 2.2624574435159395e-05,
|
| 782 |
+
"loss": 0.9442,
|
| 783 |
+
"step": 12500
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 3.9,
|
| 787 |
+
"learning_rate": 2.200557103064067e-05,
|
| 788 |
+
"loss": 0.9419,
|
| 789 |
+
"step": 12600
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 3.93,
|
| 793 |
+
"learning_rate": 2.1386567626121946e-05,
|
| 794 |
+
"loss": 0.9509,
|
| 795 |
+
"step": 12700
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 3.96,
|
| 799 |
+
"learning_rate": 2.076756422160322e-05,
|
| 800 |
+
"loss": 0.9562,
|
| 801 |
+
"step": 12800
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 3.99,
|
| 805 |
+
"learning_rate": 2.0148560817084494e-05,
|
| 806 |
+
"loss": 0.9459,
|
| 807 |
+
"step": 12900
|
| 808 |
},
|
| 809 |
{
|
| 810 |
"epoch": 4.0,
|
| 811 |
+
"eval_loss": 1.2027287483215332,
|
| 812 |
+
"eval_runtime": 100.7364,
|
| 813 |
+
"eval_samples_per_second": 35.588,
|
| 814 |
+
"eval_steps_per_second": 8.904,
|
| 815 |
+
"step": 12926
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 4.02,
|
| 819 |
+
"learning_rate": 1.952955741256577e-05,
|
| 820 |
+
"loss": 0.9155,
|
| 821 |
+
"step": 13000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 4.05,
|
| 825 |
+
"learning_rate": 1.8910554008047046e-05,
|
| 826 |
+
"loss": 0.9265,
|
| 827 |
+
"step": 13100
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 4.08,
|
| 831 |
+
"learning_rate": 1.829155060352832e-05,
|
| 832 |
+
"loss": 0.9157,
|
| 833 |
+
"step": 13200
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 4.12,
|
| 837 |
+
"learning_rate": 1.7672547199009594e-05,
|
| 838 |
+
"loss": 0.9087,
|
| 839 |
+
"step": 13300
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 4.15,
|
| 843 |
+
"learning_rate": 1.705354379449087e-05,
|
| 844 |
+
"loss": 0.9044,
|
| 845 |
+
"step": 13400
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 4.18,
|
| 849 |
+
"learning_rate": 1.6434540389972145e-05,
|
| 850 |
+
"loss": 0.9176,
|
| 851 |
+
"step": 13500
|
| 852 |
},
|
| 853 |
{
|
| 854 |
"epoch": 4.21,
|
| 855 |
+
"learning_rate": 1.581553698545342e-05,
|
| 856 |
+
"loss": 0.904,
|
| 857 |
+
"step": 13600
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 4.24,
|
| 861 |
+
"learning_rate": 1.5196533580934697e-05,
|
| 862 |
+
"loss": 0.9292,
|
| 863 |
+
"step": 13700
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 4.27,
|
| 867 |
+
"learning_rate": 1.4577530176415971e-05,
|
| 868 |
+
"loss": 0.9154,
|
| 869 |
+
"step": 13800
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 4.3,
|
| 873 |
+
"learning_rate": 1.3958526771897245e-05,
|
| 874 |
+
"loss": 0.9146,
|
| 875 |
+
"step": 13900
|
| 876 |
},
|
| 877 |
{
|
| 878 |
"epoch": 4.33,
|
| 879 |
+
"learning_rate": 1.3339523367378521e-05,
|
| 880 |
+
"loss": 0.9293,
|
| 881 |
+
"step": 14000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 4.36,
|
| 885 |
+
"learning_rate": 1.2720519962859798e-05,
|
| 886 |
+
"loss": 0.9019,
|
| 887 |
+
"step": 14100
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 4.39,
|
| 891 |
+
"learning_rate": 1.210151655834107e-05,
|
| 892 |
+
"loss": 0.902,
|
| 893 |
+
"step": 14200
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 4.43,
|
| 897 |
+
"learning_rate": 1.1482513153822347e-05,
|
| 898 |
+
"loss": 0.9085,
|
| 899 |
+
"step": 14300
|
| 900 |
},
|
| 901 |
{
|
| 902 |
"epoch": 4.46,
|
| 903 |
+
"learning_rate": 1.086350974930362e-05,
|
| 904 |
+
"loss": 0.8968,
|
| 905 |
+
"step": 14400
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 4.49,
|
| 909 |
+
"learning_rate": 1.0244506344784898e-05,
|
| 910 |
+
"loss": 0.9086,
|
| 911 |
+
"step": 14500
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 4.52,
|
| 915 |
+
"learning_rate": 9.625502940266172e-06,
|
| 916 |
+
"loss": 0.9153,
|
| 917 |
+
"step": 14600
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 4.55,
|
| 921 |
+
"learning_rate": 9.006499535747446e-06,
|
| 922 |
+
"loss": 0.9088,
|
| 923 |
+
"step": 14700
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 4.58,
|
| 927 |
+
"learning_rate": 8.387496131228722e-06,
|
| 928 |
+
"loss": 0.9106,
|
| 929 |
+
"step": 14800
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 4.61,
|
| 933 |
+
"learning_rate": 7.768492726709998e-06,
|
| 934 |
+
"loss": 0.9198,
|
| 935 |
+
"step": 14900
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 4.64,
|
| 939 |
+
"learning_rate": 7.149489322191272e-06,
|
| 940 |
+
"loss": 0.8956,
|
| 941 |
+
"step": 15000
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 4.67,
|
| 945 |
+
"learning_rate": 6.530485917672548e-06,
|
| 946 |
+
"loss": 0.8981,
|
| 947 |
+
"step": 15100
|
| 948 |
},
|
| 949 |
{
|
| 950 |
"epoch": 4.7,
|
| 951 |
+
"learning_rate": 5.9114825131538225e-06,
|
| 952 |
+
"loss": 0.9142,
|
| 953 |
+
"step": 15200
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 4.73,
|
| 957 |
+
"learning_rate": 5.2924791086350974e-06,
|
| 958 |
+
"loss": 0.9095,
|
| 959 |
+
"step": 15300
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 4.77,
|
| 963 |
+
"learning_rate": 4.673475704116373e-06,
|
| 964 |
+
"loss": 0.922,
|
| 965 |
+
"step": 15400
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 4.8,
|
| 969 |
+
"learning_rate": 4.054472299597648e-06,
|
| 970 |
+
"loss": 0.9119,
|
| 971 |
+
"step": 15500
|
| 972 |
},
|
| 973 |
{
|
| 974 |
"epoch": 4.83,
|
| 975 |
+
"learning_rate": 3.4354688950789226e-06,
|
| 976 |
+
"loss": 0.9313,
|
| 977 |
+
"step": 15600
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 4.86,
|
| 981 |
+
"learning_rate": 2.816465490560198e-06,
|
| 982 |
+
"loss": 0.9035,
|
| 983 |
+
"step": 15700
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 4.89,
|
| 987 |
+
"learning_rate": 2.1974620860414733e-06,
|
| 988 |
+
"loss": 0.9041,
|
| 989 |
+
"step": 15800
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 4.92,
|
| 993 |
+
"learning_rate": 1.5784586815227482e-06,
|
| 994 |
+
"loss": 0.9316,
|
| 995 |
+
"step": 15900
|
| 996 |
},
|
| 997 |
{
|
| 998 |
"epoch": 4.95,
|
| 999 |
+
"learning_rate": 9.594552770040236e-07,
|
| 1000 |
+
"loss": 0.9313,
|
| 1001 |
+
"step": 16000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 4.98,
|
| 1005 |
+
"learning_rate": 3.404518724852987e-07,
|
| 1006 |
+
"loss": 0.9191,
|
| 1007 |
+
"step": 16100
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
"epoch": 5.0,
|
| 1011 |
+
"eval_loss": 1.2074109315872192,
|
| 1012 |
+
"eval_runtime": 100.8821,
|
| 1013 |
+
"eval_samples_per_second": 35.537,
|
| 1014 |
+
"eval_steps_per_second": 8.892,
|
| 1015 |
+
"step": 16155
|
| 1016 |
},
|
| 1017 |
{
|
| 1018 |
"epoch": 5.0,
|
| 1019 |
+
"step": 16155,
|
| 1020 |
+
"total_flos": 1.574034355519488e+17,
|
| 1021 |
+
"train_loss": 1.0387100059157441,
|
| 1022 |
+
"train_runtime": 30224.2981,
|
| 1023 |
+
"train_samples_per_second": 8.553,
|
| 1024 |
+
"train_steps_per_second": 0.535
|
| 1025 |
}
|
| 1026 |
],
|
| 1027 |
"logging_steps": 100,
|
| 1028 |
+
"max_steps": 16155,
|
| 1029 |
"num_train_epochs": 5,
|
| 1030 |
"save_steps": 500,
|
| 1031 |
+
"total_flos": 1.574034355519488e+17,
|
| 1032 |
"trial_name": null,
|
| 1033 |
"trial_params": null
|
| 1034 |
}
|