End of training
Browse files- README.md +3 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +528 -0
README.md
CHANGED
|
@@ -4,7 +4,7 @@ base_model: google/long-t5-tglobal-xl
|
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
-
- scrolls
|
| 8 |
model-index:
|
| 9 |
- name: longt5_xl_sfd_4096_e10
|
| 10 |
results: []
|
|
@@ -15,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# longt5_xl_sfd_4096_e10
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [google/long-t5-tglobal-xl](https://huggingface.co/google/long-t5-tglobal-xl) on the scrolls dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss:
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
+
- tau/scrolls
|
| 8 |
model-index:
|
| 9 |
- name: longt5_xl_sfd_4096_e10
|
| 10 |
results: []
|
|
|
|
| 15 |
|
| 16 |
# longt5_xl_sfd_4096_e10
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [google/long-t5-tglobal-xl](https://huggingface.co/google/long-t5-tglobal-xl) on the tau/scrolls summ_screen_fd dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 2.3255
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.74,
|
| 3 |
+
"eval_loss": 2.3254711627960205,
|
| 4 |
+
"eval_runtime": 78.0006,
|
| 5 |
+
"eval_samples": 338,
|
| 6 |
+
"eval_samples_per_second": 4.333,
|
| 7 |
+
"eval_steps_per_second": 0.551,
|
| 8 |
+
"train_loss": 1.5745136559009552,
|
| 9 |
+
"train_runtime": 36489.2213,
|
| 10 |
+
"train_samples": 3673,
|
| 11 |
+
"train_samples_per_second": 1.007,
|
| 12 |
+
"train_steps_per_second": 0.004
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.74,
|
| 3 |
+
"eval_loss": 2.3254711627960205,
|
| 4 |
+
"eval_runtime": 78.0006,
|
| 5 |
+
"eval_samples": 338,
|
| 6 |
+
"eval_samples_per_second": 4.333,
|
| 7 |
+
"eval_steps_per_second": 0.551
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.74,
|
| 3 |
+
"train_loss": 1.5745136559009552,
|
| 4 |
+
"train_runtime": 36489.2213,
|
| 5 |
+
"train_samples": 3673,
|
| 6 |
+
"train_samples_per_second": 1.007,
|
| 7 |
+
"train_steps_per_second": 0.004
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.3254711627960205,
|
| 3 |
+
"best_model_checkpoint": "/home/co-ou1/rds/hpc-work/models/longt5_xl_sfd_4096/longt5_xl_sfd_4096_e10/checkpoint-28",
|
| 4 |
+
"epoch": 9.73913043478261,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 140,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.14,
|
| 13 |
+
"learning_rate": 0.001,
|
| 14 |
+
"loss": 3.2585,
|
| 15 |
+
"step": 2
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.28,
|
| 19 |
+
"learning_rate": 0.001,
|
| 20 |
+
"loss": 3.3242,
|
| 21 |
+
"step": 4
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.42,
|
| 25 |
+
"learning_rate": 0.001,
|
| 26 |
+
"loss": 3.6288,
|
| 27 |
+
"step": 6
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.56,
|
| 31 |
+
"learning_rate": 0.001,
|
| 32 |
+
"loss": 3.1948,
|
| 33 |
+
"step": 8
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.7,
|
| 37 |
+
"learning_rate": 0.001,
|
| 38 |
+
"loss": 2.9524,
|
| 39 |
+
"step": 10
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.83,
|
| 43 |
+
"learning_rate": 0.001,
|
| 44 |
+
"loss": 2.8889,
|
| 45 |
+
"step": 12
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.97,
|
| 49 |
+
"learning_rate": 0.001,
|
| 50 |
+
"loss": 3.0332,
|
| 51 |
+
"step": 14
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.97,
|
| 55 |
+
"eval_loss": 2.5423872470855713,
|
| 56 |
+
"eval_runtime": 96.2671,
|
| 57 |
+
"eval_samples_per_second": 3.511,
|
| 58 |
+
"eval_steps_per_second": 0.447,
|
| 59 |
+
"step": 14
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.11,
|
| 63 |
+
"learning_rate": 0.001,
|
| 64 |
+
"loss": 2.6074,
|
| 65 |
+
"step": 16
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.25,
|
| 69 |
+
"learning_rate": 0.001,
|
| 70 |
+
"loss": 2.6808,
|
| 71 |
+
"step": 18
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.39,
|
| 75 |
+
"learning_rate": 0.001,
|
| 76 |
+
"loss": 2.6085,
|
| 77 |
+
"step": 20
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.53,
|
| 81 |
+
"learning_rate": 0.001,
|
| 82 |
+
"loss": 2.5063,
|
| 83 |
+
"step": 22
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.67,
|
| 87 |
+
"learning_rate": 0.001,
|
| 88 |
+
"loss": 2.4614,
|
| 89 |
+
"step": 24
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.81,
|
| 93 |
+
"learning_rate": 0.001,
|
| 94 |
+
"loss": 2.4519,
|
| 95 |
+
"step": 26
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.95,
|
| 99 |
+
"learning_rate": 0.001,
|
| 100 |
+
"loss": 2.4105,
|
| 101 |
+
"step": 28
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.95,
|
| 105 |
+
"eval_loss": 2.3254711627960205,
|
| 106 |
+
"eval_runtime": 99.9342,
|
| 107 |
+
"eval_samples_per_second": 3.382,
|
| 108 |
+
"eval_steps_per_second": 0.43,
|
| 109 |
+
"step": 28
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.09,
|
| 113 |
+
"learning_rate": 0.001,
|
| 114 |
+
"loss": 2.1628,
|
| 115 |
+
"step": 30
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 2.23,
|
| 119 |
+
"learning_rate": 0.001,
|
| 120 |
+
"loss": 2.0701,
|
| 121 |
+
"step": 32
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 2.37,
|
| 125 |
+
"learning_rate": 0.001,
|
| 126 |
+
"loss": 2.0992,
|
| 127 |
+
"step": 34
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 2.5,
|
| 131 |
+
"learning_rate": 0.001,
|
| 132 |
+
"loss": 2.0401,
|
| 133 |
+
"step": 36
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 2.64,
|
| 137 |
+
"learning_rate": 0.001,
|
| 138 |
+
"loss": 2.0299,
|
| 139 |
+
"step": 38
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.78,
|
| 143 |
+
"learning_rate": 0.001,
|
| 144 |
+
"loss": 2.0812,
|
| 145 |
+
"step": 40
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 2.92,
|
| 149 |
+
"learning_rate": 0.001,
|
| 150 |
+
"loss": 2.0496,
|
| 151 |
+
"step": 42
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.99,
|
| 155 |
+
"eval_loss": 2.3419768810272217,
|
| 156 |
+
"eval_runtime": 99.9699,
|
| 157 |
+
"eval_samples_per_second": 3.381,
|
| 158 |
+
"eval_steps_per_second": 0.43,
|
| 159 |
+
"step": 43
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 3.06,
|
| 163 |
+
"learning_rate": 0.001,
|
| 164 |
+
"loss": 1.9994,
|
| 165 |
+
"step": 44
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 3.2,
|
| 169 |
+
"learning_rate": 0.001,
|
| 170 |
+
"loss": 1.7276,
|
| 171 |
+
"step": 46
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 3.34,
|
| 175 |
+
"learning_rate": 0.001,
|
| 176 |
+
"loss": 1.7639,
|
| 177 |
+
"step": 48
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.48,
|
| 181 |
+
"learning_rate": 0.001,
|
| 182 |
+
"loss": 1.7624,
|
| 183 |
+
"step": 50
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.62,
|
| 187 |
+
"learning_rate": 0.001,
|
| 188 |
+
"loss": 1.7726,
|
| 189 |
+
"step": 52
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.76,
|
| 193 |
+
"learning_rate": 0.001,
|
| 194 |
+
"loss": 1.7218,
|
| 195 |
+
"step": 54
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.9,
|
| 199 |
+
"learning_rate": 0.001,
|
| 200 |
+
"loss": 1.7473,
|
| 201 |
+
"step": 56
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.97,
|
| 205 |
+
"eval_loss": 2.352036476135254,
|
| 206 |
+
"eval_runtime": 109.1375,
|
| 207 |
+
"eval_samples_per_second": 3.097,
|
| 208 |
+
"eval_steps_per_second": 0.394,
|
| 209 |
+
"step": 57
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.03,
|
| 213 |
+
"learning_rate": 0.001,
|
| 214 |
+
"loss": 1.6586,
|
| 215 |
+
"step": 58
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 4.17,
|
| 219 |
+
"learning_rate": 0.001,
|
| 220 |
+
"loss": 1.3888,
|
| 221 |
+
"step": 60
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 4.31,
|
| 225 |
+
"learning_rate": 0.001,
|
| 226 |
+
"loss": 1.4192,
|
| 227 |
+
"step": 62
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 4.45,
|
| 231 |
+
"learning_rate": 0.001,
|
| 232 |
+
"loss": 1.4003,
|
| 233 |
+
"step": 64
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 4.59,
|
| 237 |
+
"learning_rate": 0.001,
|
| 238 |
+
"loss": 1.4405,
|
| 239 |
+
"step": 66
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 4.73,
|
| 243 |
+
"learning_rate": 0.001,
|
| 244 |
+
"loss": 1.3766,
|
| 245 |
+
"step": 68
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 4.87,
|
| 249 |
+
"learning_rate": 0.001,
|
| 250 |
+
"loss": 1.4007,
|
| 251 |
+
"step": 70
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.94,
|
| 255 |
+
"eval_loss": 2.4979982376098633,
|
| 256 |
+
"eval_runtime": 101.842,
|
| 257 |
+
"eval_samples_per_second": 3.319,
|
| 258 |
+
"eval_steps_per_second": 0.422,
|
| 259 |
+
"step": 71
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 5.01,
|
| 263 |
+
"learning_rate": 0.001,
|
| 264 |
+
"loss": 1.3547,
|
| 265 |
+
"step": 72
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 5.15,
|
| 269 |
+
"learning_rate": 0.001,
|
| 270 |
+
"loss": 1.3243,
|
| 271 |
+
"step": 74
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 5.29,
|
| 275 |
+
"learning_rate": 0.001,
|
| 276 |
+
"loss": 1.3494,
|
| 277 |
+
"step": 76
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 5.43,
|
| 281 |
+
"learning_rate": 0.001,
|
| 282 |
+
"loss": 1.3982,
|
| 283 |
+
"step": 78
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 5.57,
|
| 287 |
+
"learning_rate": 0.001,
|
| 288 |
+
"loss": 1.3294,
|
| 289 |
+
"step": 80
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 5.7,
|
| 293 |
+
"learning_rate": 0.001,
|
| 294 |
+
"loss": 1.404,
|
| 295 |
+
"step": 82
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 5.84,
|
| 299 |
+
"learning_rate": 0.001,
|
| 300 |
+
"loss": 1.371,
|
| 301 |
+
"step": 84
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 5.98,
|
| 305 |
+
"learning_rate": 0.001,
|
| 306 |
+
"loss": 1.3809,
|
| 307 |
+
"step": 86
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 5.98,
|
| 311 |
+
"eval_loss": 2.4785053730010986,
|
| 312 |
+
"eval_runtime": 78.2069,
|
| 313 |
+
"eval_samples_per_second": 4.322,
|
| 314 |
+
"eval_steps_per_second": 0.55,
|
| 315 |
+
"step": 86
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 6.12,
|
| 319 |
+
"learning_rate": 0.001,
|
| 320 |
+
"loss": 1.0798,
|
| 321 |
+
"step": 88
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 6.26,
|
| 325 |
+
"learning_rate": 0.001,
|
| 326 |
+
"loss": 1.0476,
|
| 327 |
+
"step": 90
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 6.4,
|
| 331 |
+
"learning_rate": 0.001,
|
| 332 |
+
"loss": 1.111,
|
| 333 |
+
"step": 92
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 6.54,
|
| 337 |
+
"learning_rate": 0.001,
|
| 338 |
+
"loss": 1.0734,
|
| 339 |
+
"step": 94
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 6.68,
|
| 343 |
+
"learning_rate": 0.001,
|
| 344 |
+
"loss": 1.0563,
|
| 345 |
+
"step": 96
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 6.82,
|
| 349 |
+
"learning_rate": 0.001,
|
| 350 |
+
"loss": 1.1215,
|
| 351 |
+
"step": 98
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 6.96,
|
| 355 |
+
"learning_rate": 0.001,
|
| 356 |
+
"loss": 1.1153,
|
| 357 |
+
"step": 100
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 6.96,
|
| 361 |
+
"eval_loss": 2.732642650604248,
|
| 362 |
+
"eval_runtime": 78.0611,
|
| 363 |
+
"eval_samples_per_second": 4.33,
|
| 364 |
+
"eval_steps_per_second": 0.551,
|
| 365 |
+
"step": 100
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 7.1,
|
| 369 |
+
"learning_rate": 0.001,
|
| 370 |
+
"loss": 0.9032,
|
| 371 |
+
"step": 102
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 7.23,
|
| 375 |
+
"learning_rate": 0.001,
|
| 376 |
+
"loss": 0.8517,
|
| 377 |
+
"step": 104
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 7.37,
|
| 381 |
+
"learning_rate": 0.001,
|
| 382 |
+
"loss": 0.8711,
|
| 383 |
+
"step": 106
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 7.51,
|
| 387 |
+
"learning_rate": 0.001,
|
| 388 |
+
"loss": 0.8849,
|
| 389 |
+
"step": 108
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 7.65,
|
| 393 |
+
"learning_rate": 0.001,
|
| 394 |
+
"loss": 0.898,
|
| 395 |
+
"step": 110
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 7.79,
|
| 399 |
+
"learning_rate": 0.001,
|
| 400 |
+
"loss": 0.9153,
|
| 401 |
+
"step": 112
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.93,
|
| 405 |
+
"learning_rate": 0.001,
|
| 406 |
+
"loss": 0.9129,
|
| 407 |
+
"step": 114
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 8.0,
|
| 411 |
+
"eval_loss": 2.9232447147369385,
|
| 412 |
+
"eval_runtime": 78.4249,
|
| 413 |
+
"eval_samples_per_second": 4.31,
|
| 414 |
+
"eval_steps_per_second": 0.548,
|
| 415 |
+
"step": 115
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 8.07,
|
| 419 |
+
"learning_rate": 0.001,
|
| 420 |
+
"loss": 0.7377,
|
| 421 |
+
"step": 116
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 8.21,
|
| 425 |
+
"learning_rate": 0.001,
|
| 426 |
+
"loss": 0.6558,
|
| 427 |
+
"step": 118
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 8.35,
|
| 431 |
+
"learning_rate": 0.001,
|
| 432 |
+
"loss": 0.7047,
|
| 433 |
+
"step": 120
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 8.49,
|
| 437 |
+
"learning_rate": 0.001,
|
| 438 |
+
"loss": 0.7382,
|
| 439 |
+
"step": 122
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 8.63,
|
| 443 |
+
"learning_rate": 0.001,
|
| 444 |
+
"loss": 0.6919,
|
| 445 |
+
"step": 124
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 8.77,
|
| 449 |
+
"learning_rate": 0.001,
|
| 450 |
+
"loss": 0.7257,
|
| 451 |
+
"step": 126
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 8.9,
|
| 455 |
+
"learning_rate": 0.001,
|
| 456 |
+
"loss": 0.7118,
|
| 457 |
+
"step": 128
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 8.97,
|
| 461 |
+
"eval_loss": 3.047579526901245,
|
| 462 |
+
"eval_runtime": 78.168,
|
| 463 |
+
"eval_samples_per_second": 4.324,
|
| 464 |
+
"eval_steps_per_second": 0.55,
|
| 465 |
+
"step": 129
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 9.04,
|
| 469 |
+
"learning_rate": 0.001,
|
| 470 |
+
"loss": 0.6401,
|
| 471 |
+
"step": 130
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 9.18,
|
| 475 |
+
"learning_rate": 0.001,
|
| 476 |
+
"loss": 0.5032,
|
| 477 |
+
"step": 132
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 9.32,
|
| 481 |
+
"learning_rate": 0.001,
|
| 482 |
+
"loss": 0.548,
|
| 483 |
+
"step": 134
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 9.46,
|
| 487 |
+
"learning_rate": 0.001,
|
| 488 |
+
"loss": 0.5218,
|
| 489 |
+
"step": 136
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 9.6,
|
| 493 |
+
"learning_rate": 0.001,
|
| 494 |
+
"loss": 0.5744,
|
| 495 |
+
"step": 138
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 9.74,
|
| 499 |
+
"learning_rate": 0.001,
|
| 500 |
+
"loss": 0.5883,
|
| 501 |
+
"step": 140
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 9.74,
|
| 505 |
+
"eval_loss": 3.3644142150878906,
|
| 506 |
+
"eval_runtime": 78.3345,
|
| 507 |
+
"eval_samples_per_second": 4.315,
|
| 508 |
+
"eval_steps_per_second": 0.549,
|
| 509 |
+
"step": 140
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 9.74,
|
| 513 |
+
"step": 140,
|
| 514 |
+
"total_flos": 2.447850236380324e+18,
|
| 515 |
+
"train_loss": 1.5745136559009552,
|
| 516 |
+
"train_runtime": 36489.2213,
|
| 517 |
+
"train_samples_per_second": 1.007,
|
| 518 |
+
"train_steps_per_second": 0.004
|
| 519 |
+
}
|
| 520 |
+
],
|
| 521 |
+
"logging_steps": 2,
|
| 522 |
+
"max_steps": 140,
|
| 523 |
+
"num_train_epochs": 10,
|
| 524 |
+
"save_steps": 500,
|
| 525 |
+
"total_flos": 2.447850236380324e+18,
|
| 526 |
+
"trial_name": null,
|
| 527 |
+
"trial_params": null
|
| 528 |
+
}
|