End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +661 -0
README.md
CHANGED
|
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [t5-3b](https://huggingface.co/t5-3b) on the squad dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [t5-3b](https://huggingface.co/t5-3b) on the squad dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.2233
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.2233106642961502,
|
| 4 |
+
"eval_runtime": 133.8529,
|
| 5 |
+
"eval_samples": 10570,
|
| 6 |
+
"eval_samples_per_second": 78.967,
|
| 7 |
+
"eval_steps_per_second": 9.877,
|
| 8 |
+
"train_loss": 0.1461177020974559,
|
| 9 |
+
"train_runtime": 16425.358,
|
| 10 |
+
"train_samples": 87599,
|
| 11 |
+
"train_samples_per_second": 15.999,
|
| 12 |
+
"train_steps_per_second": 0.062
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.2233106642961502,
|
| 4 |
+
"eval_runtime": 133.8529,
|
| 5 |
+
"eval_samples": 10570,
|
| 6 |
+
"eval_samples_per_second": 78.967,
|
| 7 |
+
"eval_steps_per_second": 9.877
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 0.1461177020974559,
|
| 4 |
+
"train_runtime": 16425.358,
|
| 5 |
+
"train_samples": 87599,
|
| 6 |
+
"train_samples_per_second": 15.999,
|
| 7 |
+
"train_steps_per_second": 0.062
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,661 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.2233106642961502,
|
| 3 |
+
"best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/question_gen/t5_3b_epoch_3_qa/checkpoint-342",
|
| 4 |
+
"epoch": 2.9983561643835617,
|
| 5 |
+
"global_step": 1026,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 9.902534113060428e-05,
|
| 13 |
+
"loss": 0.2782,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.06,
|
| 18 |
+
"learning_rate": 9.805068226120859e-05,
|
| 19 |
+
"loss": 0.221,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.09,
|
| 24 |
+
"learning_rate": 9.707602339181286e-05,
|
| 25 |
+
"loss": 0.2068,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.12,
|
| 30 |
+
"learning_rate": 9.610136452241715e-05,
|
| 31 |
+
"loss": 0.2119,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.15,
|
| 36 |
+
"learning_rate": 9.512670565302145e-05,
|
| 37 |
+
"loss": 0.2116,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.18,
|
| 42 |
+
"learning_rate": 9.415204678362574e-05,
|
| 43 |
+
"loss": 0.2104,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2,
|
| 48 |
+
"learning_rate": 9.317738791423001e-05,
|
| 49 |
+
"loss": 0.2137,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.23,
|
| 54 |
+
"learning_rate": 9.220272904483432e-05,
|
| 55 |
+
"loss": 0.2157,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.26,
|
| 60 |
+
"learning_rate": 9.12280701754386e-05,
|
| 61 |
+
"loss": 0.2052,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.29,
|
| 66 |
+
"learning_rate": 9.025341130604289e-05,
|
| 67 |
+
"loss": 0.2062,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.32,
|
| 72 |
+
"learning_rate": 8.927875243664718e-05,
|
| 73 |
+
"loss": 0.2091,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.35,
|
| 78 |
+
"learning_rate": 8.830409356725147e-05,
|
| 79 |
+
"loss": 0.2161,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.38,
|
| 84 |
+
"learning_rate": 8.732943469785574e-05,
|
| 85 |
+
"loss": 0.2051,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.41,
|
| 90 |
+
"learning_rate": 8.635477582846005e-05,
|
| 91 |
+
"loss": 0.2112,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.44,
|
| 96 |
+
"learning_rate": 8.538011695906433e-05,
|
| 97 |
+
"loss": 0.1951,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.47,
|
| 102 |
+
"learning_rate": 8.440545808966862e-05,
|
| 103 |
+
"loss": 0.2062,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.5,
|
| 108 |
+
"learning_rate": 8.343079922027291e-05,
|
| 109 |
+
"loss": 0.2046,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.53,
|
| 114 |
+
"learning_rate": 8.24561403508772e-05,
|
| 115 |
+
"loss": 0.1999,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.56,
|
| 120 |
+
"learning_rate": 8.148148148148148e-05,
|
| 121 |
+
"loss": 0.2047,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.58,
|
| 126 |
+
"learning_rate": 8.050682261208578e-05,
|
| 127 |
+
"loss": 0.2042,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.61,
|
| 132 |
+
"learning_rate": 7.953216374269006e-05,
|
| 133 |
+
"loss": 0.1995,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.64,
|
| 138 |
+
"learning_rate": 7.855750487329435e-05,
|
| 139 |
+
"loss": 0.1926,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.67,
|
| 144 |
+
"learning_rate": 7.758284600389864e-05,
|
| 145 |
+
"loss": 0.204,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.7,
|
| 150 |
+
"learning_rate": 7.660818713450293e-05,
|
| 151 |
+
"loss": 0.1925,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.73,
|
| 156 |
+
"learning_rate": 7.563352826510721e-05,
|
| 157 |
+
"loss": 0.2017,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.76,
|
| 162 |
+
"learning_rate": 7.465886939571151e-05,
|
| 163 |
+
"loss": 0.1984,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.79,
|
| 168 |
+
"learning_rate": 7.368421052631579e-05,
|
| 169 |
+
"loss": 0.1941,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.82,
|
| 174 |
+
"learning_rate": 7.270955165692008e-05,
|
| 175 |
+
"loss": 0.1932,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.85,
|
| 180 |
+
"learning_rate": 7.173489278752437e-05,
|
| 181 |
+
"loss": 0.1979,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.88,
|
| 186 |
+
"learning_rate": 7.076023391812866e-05,
|
| 187 |
+
"loss": 0.1994,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.91,
|
| 192 |
+
"learning_rate": 6.978557504873294e-05,
|
| 193 |
+
"loss": 0.1935,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.94,
|
| 198 |
+
"learning_rate": 6.881091617933724e-05,
|
| 199 |
+
"loss": 0.1951,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.96,
|
| 204 |
+
"learning_rate": 6.783625730994152e-05,
|
| 205 |
+
"loss": 0.1964,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.99,
|
| 210 |
+
"learning_rate": 6.686159844054581e-05,
|
| 211 |
+
"loss": 0.2003,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.0,
|
| 216 |
+
"eval_loss": 0.2233106642961502,
|
| 217 |
+
"eval_runtime": 133.4707,
|
| 218 |
+
"eval_samples_per_second": 79.193,
|
| 219 |
+
"eval_steps_per_second": 9.905,
|
| 220 |
+
"step": 342
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.02,
|
| 224 |
+
"learning_rate": 6.58869395711501e-05,
|
| 225 |
+
"loss": 0.1463,
|
| 226 |
+
"step": 350
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.05,
|
| 230 |
+
"learning_rate": 6.49122807017544e-05,
|
| 231 |
+
"loss": 0.1331,
|
| 232 |
+
"step": 360
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 1.08,
|
| 236 |
+
"learning_rate": 6.393762183235867e-05,
|
| 237 |
+
"loss": 0.1401,
|
| 238 |
+
"step": 370
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.11,
|
| 242 |
+
"learning_rate": 6.296296296296296e-05,
|
| 243 |
+
"loss": 0.1359,
|
| 244 |
+
"step": 380
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.14,
|
| 248 |
+
"learning_rate": 6.198830409356725e-05,
|
| 249 |
+
"loss": 0.1372,
|
| 250 |
+
"step": 390
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.17,
|
| 254 |
+
"learning_rate": 6.101364522417154e-05,
|
| 255 |
+
"loss": 0.1361,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.2,
|
| 260 |
+
"learning_rate": 6.003898635477583e-05,
|
| 261 |
+
"loss": 0.1408,
|
| 262 |
+
"step": 410
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.23,
|
| 266 |
+
"learning_rate": 5.9064327485380125e-05,
|
| 267 |
+
"loss": 0.1382,
|
| 268 |
+
"step": 420
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.26,
|
| 272 |
+
"learning_rate": 5.80896686159844e-05,
|
| 273 |
+
"loss": 0.1316,
|
| 274 |
+
"step": 430
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.29,
|
| 278 |
+
"learning_rate": 5.71150097465887e-05,
|
| 279 |
+
"loss": 0.1416,
|
| 280 |
+
"step": 440
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 1.32,
|
| 284 |
+
"learning_rate": 5.6140350877192984e-05,
|
| 285 |
+
"loss": 0.1371,
|
| 286 |
+
"step": 450
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.34,
|
| 290 |
+
"learning_rate": 5.5165692007797275e-05,
|
| 291 |
+
"loss": 0.1366,
|
| 292 |
+
"step": 460
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.37,
|
| 296 |
+
"learning_rate": 5.419103313840156e-05,
|
| 297 |
+
"loss": 0.1359,
|
| 298 |
+
"step": 470
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.4,
|
| 302 |
+
"learning_rate": 5.3216374269005856e-05,
|
| 303 |
+
"loss": 0.1395,
|
| 304 |
+
"step": 480
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.43,
|
| 308 |
+
"learning_rate": 5.2241715399610133e-05,
|
| 309 |
+
"loss": 0.1353,
|
| 310 |
+
"step": 490
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.46,
|
| 314 |
+
"learning_rate": 5.126705653021443e-05,
|
| 315 |
+
"loss": 0.1369,
|
| 316 |
+
"step": 500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.49,
|
| 320 |
+
"learning_rate": 5.0292397660818715e-05,
|
| 321 |
+
"loss": 0.1296,
|
| 322 |
+
"step": 510
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.52,
|
| 326 |
+
"learning_rate": 4.9317738791423e-05,
|
| 327 |
+
"loss": 0.1323,
|
| 328 |
+
"step": 520
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.55,
|
| 332 |
+
"learning_rate": 4.834307992202729e-05,
|
| 333 |
+
"loss": 0.1363,
|
| 334 |
+
"step": 530
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.58,
|
| 338 |
+
"learning_rate": 4.736842105263158e-05,
|
| 339 |
+
"loss": 0.1492,
|
| 340 |
+
"step": 540
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.61,
|
| 344 |
+
"learning_rate": 4.6393762183235865e-05,
|
| 345 |
+
"loss": 0.1355,
|
| 346 |
+
"step": 550
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.64,
|
| 350 |
+
"learning_rate": 4.5419103313840156e-05,
|
| 351 |
+
"loss": 0.134,
|
| 352 |
+
"step": 560
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.67,
|
| 356 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 357 |
+
"loss": 0.1385,
|
| 358 |
+
"step": 570
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 1.69,
|
| 362 |
+
"learning_rate": 4.346978557504873e-05,
|
| 363 |
+
"loss": 0.1301,
|
| 364 |
+
"step": 580
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 1.72,
|
| 368 |
+
"learning_rate": 4.249512670565302e-05,
|
| 369 |
+
"loss": 0.1442,
|
| 370 |
+
"step": 590
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.75,
|
| 374 |
+
"learning_rate": 4.152046783625731e-05,
|
| 375 |
+
"loss": 0.135,
|
| 376 |
+
"step": 600
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.78,
|
| 380 |
+
"learning_rate": 4.0545808966861596e-05,
|
| 381 |
+
"loss": 0.1278,
|
| 382 |
+
"step": 610
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.81,
|
| 386 |
+
"learning_rate": 3.957115009746589e-05,
|
| 387 |
+
"loss": 0.1389,
|
| 388 |
+
"step": 620
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.84,
|
| 392 |
+
"learning_rate": 3.859649122807018e-05,
|
| 393 |
+
"loss": 0.1386,
|
| 394 |
+
"step": 630
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.87,
|
| 398 |
+
"learning_rate": 3.762183235867446e-05,
|
| 399 |
+
"loss": 0.1313,
|
| 400 |
+
"step": 640
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 1.9,
|
| 404 |
+
"learning_rate": 3.664717348927875e-05,
|
| 405 |
+
"loss": 0.1325,
|
| 406 |
+
"step": 650
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.93,
|
| 410 |
+
"learning_rate": 3.5672514619883044e-05,
|
| 411 |
+
"loss": 0.1315,
|
| 412 |
+
"step": 660
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 1.96,
|
| 416 |
+
"learning_rate": 3.469785575048733e-05,
|
| 417 |
+
"loss": 0.1344,
|
| 418 |
+
"step": 670
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.99,
|
| 422 |
+
"learning_rate": 3.372319688109162e-05,
|
| 423 |
+
"loss": 0.1331,
|
| 424 |
+
"step": 680
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.0,
|
| 428 |
+
"eval_loss": 0.2396751344203949,
|
| 429 |
+
"eval_runtime": 133.3515,
|
| 430 |
+
"eval_samples_per_second": 79.264,
|
| 431 |
+
"eval_steps_per_second": 9.914,
|
| 432 |
+
"step": 684
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.02,
|
| 436 |
+
"learning_rate": 3.274853801169591e-05,
|
| 437 |
+
"loss": 0.1059,
|
| 438 |
+
"step": 690
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.05,
|
| 442 |
+
"learning_rate": 3.1773879142300193e-05,
|
| 443 |
+
"loss": 0.1006,
|
| 444 |
+
"step": 700
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.07,
|
| 448 |
+
"learning_rate": 3.0799220272904484e-05,
|
| 449 |
+
"loss": 0.0941,
|
| 450 |
+
"step": 710
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 2.1,
|
| 454 |
+
"learning_rate": 2.9824561403508772e-05,
|
| 455 |
+
"loss": 0.1033,
|
| 456 |
+
"step": 720
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 2.13,
|
| 460 |
+
"learning_rate": 2.884990253411306e-05,
|
| 461 |
+
"loss": 0.0978,
|
| 462 |
+
"step": 730
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.16,
|
| 466 |
+
"learning_rate": 2.787524366471735e-05,
|
| 467 |
+
"loss": 0.0983,
|
| 468 |
+
"step": 740
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 2.19,
|
| 472 |
+
"learning_rate": 2.6900584795321637e-05,
|
| 473 |
+
"loss": 0.0987,
|
| 474 |
+
"step": 750
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.22,
|
| 478 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 479 |
+
"loss": 0.0985,
|
| 480 |
+
"step": 760
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.25,
|
| 484 |
+
"learning_rate": 2.4951267056530216e-05,
|
| 485 |
+
"loss": 0.0974,
|
| 486 |
+
"step": 770
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 2.28,
|
| 490 |
+
"learning_rate": 2.3976608187134503e-05,
|
| 491 |
+
"loss": 0.1001,
|
| 492 |
+
"step": 780
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.31,
|
| 496 |
+
"learning_rate": 2.300194931773879e-05,
|
| 497 |
+
"loss": 0.0961,
|
| 498 |
+
"step": 790
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 2.34,
|
| 502 |
+
"learning_rate": 2.2027290448343078e-05,
|
| 503 |
+
"loss": 0.0958,
|
| 504 |
+
"step": 800
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.37,
|
| 508 |
+
"learning_rate": 2.105263157894737e-05,
|
| 509 |
+
"loss": 0.0975,
|
| 510 |
+
"step": 810
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.4,
|
| 514 |
+
"learning_rate": 2.0077972709551656e-05,
|
| 515 |
+
"loss": 0.0955,
|
| 516 |
+
"step": 820
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 2.43,
|
| 520 |
+
"learning_rate": 1.9103313840155944e-05,
|
| 521 |
+
"loss": 0.0927,
|
| 522 |
+
"step": 830
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.45,
|
| 526 |
+
"learning_rate": 1.8128654970760235e-05,
|
| 527 |
+
"loss": 0.0958,
|
| 528 |
+
"step": 840
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 2.48,
|
| 532 |
+
"learning_rate": 1.7153996101364522e-05,
|
| 533 |
+
"loss": 0.0979,
|
| 534 |
+
"step": 850
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.51,
|
| 538 |
+
"learning_rate": 1.617933723196881e-05,
|
| 539 |
+
"loss": 0.1,
|
| 540 |
+
"step": 860
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.54,
|
| 544 |
+
"learning_rate": 1.5204678362573099e-05,
|
| 545 |
+
"loss": 0.0965,
|
| 546 |
+
"step": 870
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.57,
|
| 550 |
+
"learning_rate": 1.4230019493177388e-05,
|
| 551 |
+
"loss": 0.098,
|
| 552 |
+
"step": 880
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.6,
|
| 556 |
+
"learning_rate": 1.3255360623781677e-05,
|
| 557 |
+
"loss": 0.0981,
|
| 558 |
+
"step": 890
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 2.63,
|
| 562 |
+
"learning_rate": 1.2280701754385964e-05,
|
| 563 |
+
"loss": 0.1024,
|
| 564 |
+
"step": 900
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.66,
|
| 568 |
+
"learning_rate": 1.1306042884990253e-05,
|
| 569 |
+
"loss": 0.0962,
|
| 570 |
+
"step": 910
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 2.69,
|
| 574 |
+
"learning_rate": 1.0331384015594543e-05,
|
| 575 |
+
"loss": 0.0993,
|
| 576 |
+
"step": 920
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.72,
|
| 580 |
+
"learning_rate": 9.35672514619883e-06,
|
| 581 |
+
"loss": 0.0932,
|
| 582 |
+
"step": 930
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.75,
|
| 586 |
+
"learning_rate": 8.38206627680312e-06,
|
| 587 |
+
"loss": 0.0929,
|
| 588 |
+
"step": 940
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 2.78,
|
| 592 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 593 |
+
"loss": 0.0881,
|
| 594 |
+
"step": 950
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.81,
|
| 598 |
+
"learning_rate": 6.432748538011696e-06,
|
| 599 |
+
"loss": 0.0923,
|
| 600 |
+
"step": 960
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.83,
|
| 604 |
+
"learning_rate": 5.458089668615984e-06,
|
| 605 |
+
"loss": 0.0935,
|
| 606 |
+
"step": 970
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.86,
|
| 610 |
+
"learning_rate": 4.483430799220273e-06,
|
| 611 |
+
"loss": 0.1015,
|
| 612 |
+
"step": 980
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 2.89,
|
| 616 |
+
"learning_rate": 3.5087719298245615e-06,
|
| 617 |
+
"loss": 0.0939,
|
| 618 |
+
"step": 990
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.92,
|
| 622 |
+
"learning_rate": 2.5341130604288498e-06,
|
| 623 |
+
"loss": 0.093,
|
| 624 |
+
"step": 1000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.95,
|
| 628 |
+
"learning_rate": 1.5594541910331385e-06,
|
| 629 |
+
"loss": 0.0975,
|
| 630 |
+
"step": 1010
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 2.98,
|
| 634 |
+
"learning_rate": 5.847953216374269e-07,
|
| 635 |
+
"loss": 0.0971,
|
| 636 |
+
"step": 1020
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 3.0,
|
| 640 |
+
"eval_loss": 0.26710498332977295,
|
| 641 |
+
"eval_runtime": 133.9576,
|
| 642 |
+
"eval_samples_per_second": 78.906,
|
| 643 |
+
"eval_steps_per_second": 9.869,
|
| 644 |
+
"step": 1026
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 3.0,
|
| 648 |
+
"step": 1026,
|
| 649 |
+
"total_flos": 1.4063349020358083e+18,
|
| 650 |
+
"train_loss": 0.1461177020974559,
|
| 651 |
+
"train_runtime": 16425.358,
|
| 652 |
+
"train_samples_per_second": 15.999,
|
| 653 |
+
"train_steps_per_second": 0.062
|
| 654 |
+
}
|
| 655 |
+
],
|
| 656 |
+
"max_steps": 1026,
|
| 657 |
+
"num_train_epochs": 3,
|
| 658 |
+
"total_flos": 1.4063349020358083e+18,
|
| 659 |
+
"trial_name": null,
|
| 660 |
+
"trial_params": null
|
| 661 |
+
}
|