End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 17.7,
|
| 3 |
+
"eval_accuracy": 0.05156855736415957,
|
| 4 |
+
"eval_loss": 7.338886260986328,
|
| 5 |
+
"eval_runtime": 180.6703,
|
| 6 |
+
"eval_samples": 24055,
|
| 7 |
+
"eval_samples_per_second": 133.143,
|
| 8 |
+
"eval_steps_per_second": 4.162,
|
| 9 |
+
"perplexity": 1538.997117682623,
|
| 10 |
+
"train_loss": 7.482659651812385,
|
| 11 |
+
"train_runtime": 11122.8848,
|
| 12 |
+
"train_samples": 24910,
|
| 13 |
+
"train_samples_per_second": 223.953,
|
| 14 |
+
"train_steps_per_second": 0.432
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 17.7,
|
| 3 |
+
"eval_accuracy": 0.05156855736415957,
|
| 4 |
+
"eval_loss": 7.338886260986328,
|
| 5 |
+
"eval_runtime": 180.6703,
|
| 6 |
+
"eval_samples": 24055,
|
| 7 |
+
"eval_samples_per_second": 133.143,
|
| 8 |
+
"eval_steps_per_second": 4.162,
|
| 9 |
+
"perplexity": 1538.997117682623
|
| 10 |
+
}
|
runs/Feb16_23-05-57_monica.sm.unibocconi.it/events.out.tfevents.1676596535.monica.sm.unibocconi.it.1910604.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:540f2890224e03c7ec1c499bf032a71d0428638e5c8be8c678557ed5e98b2c3b
|
| 3 |
+
size 363
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 17.7,
|
| 3 |
+
"train_loss": 7.482659651812385,
|
| 4 |
+
"train_runtime": 11122.8848,
|
| 5 |
+
"train_samples": 24910,
|
| 6 |
+
"train_samples_per_second": 223.953,
|
| 7 |
+
"train_steps_per_second": 0.432
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,688 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 7.342555999755859,
|
| 3 |
+
"best_model_checkpoint": "/data1/attanasiog/babylm/roberta-tiny-8l-10M/checkpoint-700",
|
| 4 |
+
"epoch": 17.698779704560053,
|
| 5 |
+
"global_step": 850,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.21,
|
| 12 |
+
"learning_rate": 8e-05,
|
| 13 |
+
"loss": 10.2998,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.41,
|
| 18 |
+
"learning_rate": 0.00016,
|
| 19 |
+
"loss": 8.9979,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.62,
|
| 24 |
+
"learning_rate": 0.00024,
|
| 25 |
+
"loss": 7.8015,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.82,
|
| 30 |
+
"learning_rate": 0.00032,
|
| 31 |
+
"loss": 7.3376,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.04,
|
| 36 |
+
"learning_rate": 0.0004,
|
| 37 |
+
"loss": 7.8102,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.04,
|
| 42 |
+
"eval_accuracy": 0.05136765891155645,
|
| 43 |
+
"eval_loss": 7.374657154083252,
|
| 44 |
+
"eval_runtime": 180.8913,
|
| 45 |
+
"eval_samples_per_second": 132.98,
|
| 46 |
+
"eval_steps_per_second": 4.157,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.25,
|
| 51 |
+
"learning_rate": 0.000399995625676045,
|
| 52 |
+
"loss": 7.3419,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.45,
|
| 57 |
+
"learning_rate": 0.0003999825028955268,
|
| 58 |
+
"loss": 8.1652,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.66,
|
| 63 |
+
"learning_rate": 0.0003999606322324786,
|
| 64 |
+
"loss": 8.4182,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.86,
|
| 69 |
+
"learning_rate": 0.0003999300146435939,
|
| 70 |
+
"loss": 7.3249,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 2.08,
|
| 75 |
+
"learning_rate": 0.00039989065146818525,
|
| 76 |
+
"loss": 7.805,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.08,
|
| 81 |
+
"eval_accuracy": 0.051684268514440884,
|
| 82 |
+
"eval_loss": 7.369903087615967,
|
| 83 |
+
"eval_runtime": 181.0798,
|
| 84 |
+
"eval_samples_per_second": 132.842,
|
| 85 |
+
"eval_steps_per_second": 4.153,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.29,
|
| 90 |
+
"learning_rate": 0.0003998425444281255,
|
| 91 |
+
"loss": 7.3101,
|
| 92 |
+
"step": 110
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.49,
|
| 96 |
+
"learning_rate": 0.00039978569562777234,
|
| 97 |
+
"loss": 7.3232,
|
| 98 |
+
"step": 120
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.7,
|
| 102 |
+
"learning_rate": 0.0003997201075538765,
|
| 103 |
+
"loss": 7.3073,
|
| 104 |
+
"step": 130
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.9,
|
| 108 |
+
"learning_rate": 0.0003996457830754729,
|
| 109 |
+
"loss": 7.3236,
|
| 110 |
+
"step": 140
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.12,
|
| 114 |
+
"learning_rate": 0.00039956272544375493,
|
| 115 |
+
"loss": 7.7907,
|
| 116 |
+
"step": 150
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.12,
|
| 120 |
+
"eval_accuracy": 0.05174263561361906,
|
| 121 |
+
"eval_loss": 7.35952091217041,
|
| 122 |
+
"eval_runtime": 180.7769,
|
| 123 |
+
"eval_samples_per_second": 133.065,
|
| 124 |
+
"eval_steps_per_second": 4.16,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.33,
|
| 129 |
+
"learning_rate": 0.00039947093829193245,
|
| 130 |
+
"loss": 7.2981,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.53,
|
| 135 |
+
"learning_rate": 0.00039937042563507283,
|
| 136 |
+
"loss": 7.3259,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 3.74,
|
| 141 |
+
"learning_rate": 0.00039926119186992537,
|
| 142 |
+
"loss": 7.3352,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.95,
|
| 147 |
+
"learning_rate": 0.0003991432417747288,
|
| 148 |
+
"loss": 7.3069,
|
| 149 |
+
"step": 190
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 4.16,
|
| 153 |
+
"learning_rate": 0.0003990165805090023,
|
| 154 |
+
"loss": 7.7838,
|
| 155 |
+
"step": 200
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 4.16,
|
| 159 |
+
"eval_accuracy": 0.05138188801155976,
|
| 160 |
+
"eval_loss": 7.361721992492676,
|
| 161 |
+
"eval_runtime": 180.6907,
|
| 162 |
+
"eval_samples_per_second": 133.128,
|
| 163 |
+
"eval_steps_per_second": 4.162,
|
| 164 |
+
"step": 200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.37,
|
| 168 |
+
"learning_rate": 0.00039888121361332003,
|
| 169 |
+
"loss": 7.3066,
|
| 170 |
+
"step": 210
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 4.58,
|
| 174 |
+
"learning_rate": 0.0003987371470090686,
|
| 175 |
+
"loss": 7.3237,
|
| 176 |
+
"step": 220
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 4.78,
|
| 180 |
+
"learning_rate": 0.00039858438699818784,
|
| 181 |
+
"loss": 7.3209,
|
| 182 |
+
"step": 230
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 4.99,
|
| 186 |
+
"learning_rate": 0.0003984229402628956,
|
| 187 |
+
"loss": 7.3024,
|
| 188 |
+
"step": 240
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 5.21,
|
| 192 |
+
"learning_rate": 0.00039825281386539503,
|
| 193 |
+
"loss": 7.7706,
|
| 194 |
+
"step": 250
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.21,
|
| 198 |
+
"eval_accuracy": 0.05140231728427503,
|
| 199 |
+
"eval_loss": 7.358623504638672,
|
| 200 |
+
"eval_runtime": 180.7786,
|
| 201 |
+
"eval_samples_per_second": 133.063,
|
| 202 |
+
"eval_steps_per_second": 4.16,
|
| 203 |
+
"step": 250
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.41,
|
| 207 |
+
"learning_rate": 0.000398074015247566,
|
| 208 |
+
"loss": 7.3135,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 5.62,
|
| 213 |
+
"learning_rate": 0.0003978865522306392,
|
| 214 |
+
"loss": 7.3003,
|
| 215 |
+
"step": 270
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.82,
|
| 219 |
+
"learning_rate": 0.0003976904330148543,
|
| 220 |
+
"loss": 7.3159,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 6.04,
|
| 225 |
+
"learning_rate": 0.00039748566617910113,
|
| 226 |
+
"loss": 7.7967,
|
| 227 |
+
"step": 290
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 6.25,
|
| 231 |
+
"learning_rate": 0.0003972722606805445,
|
| 232 |
+
"loss": 7.2933,
|
| 233 |
+
"step": 300
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 6.25,
|
| 237 |
+
"eval_accuracy": 0.05126180317018771,
|
| 238 |
+
"eval_loss": 7.356584548950195,
|
| 239 |
+
"eval_runtime": 180.7497,
|
| 240 |
+
"eval_samples_per_second": 133.085,
|
| 241 |
+
"eval_steps_per_second": 4.16,
|
| 242 |
+
"step": 300
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 6.45,
|
| 246 |
+
"learning_rate": 0.00039705022585423216,
|
| 247 |
+
"loss": 7.3163,
|
| 248 |
+
"step": 310
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 6.66,
|
| 252 |
+
"learning_rate": 0.0003968195714126868,
|
| 253 |
+
"loss": 7.2904,
|
| 254 |
+
"step": 320
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 6.86,
|
| 258 |
+
"learning_rate": 0.00039658030744548075,
|
| 259 |
+
"loss": 7.3045,
|
| 260 |
+
"step": 330
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 7.08,
|
| 264 |
+
"learning_rate": 0.0003963324444187952,
|
| 265 |
+
"loss": 7.7849,
|
| 266 |
+
"step": 340
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 7.29,
|
| 270 |
+
"learning_rate": 0.0003960759931749619,
|
| 271 |
+
"loss": 7.2932,
|
| 272 |
+
"step": 350
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 7.29,
|
| 276 |
+
"eval_accuracy": 0.05161072401384023,
|
| 277 |
+
"eval_loss": 7.3526611328125,
|
| 278 |
+
"eval_runtime": 180.6553,
|
| 279 |
+
"eval_samples_per_second": 133.154,
|
| 280 |
+
"eval_steps_per_second": 4.163,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 7.49,
|
| 285 |
+
"learning_rate": 0.00039581096493198893,
|
| 286 |
+
"loss": 7.3057,
|
| 287 |
+
"step": 360
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 7.7,
|
| 291 |
+
"learning_rate": 0.0003955373712830703,
|
| 292 |
+
"loss": 7.3002,
|
| 293 |
+
"step": 370
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 7.9,
|
| 297 |
+
"learning_rate": 0.00039525522419607854,
|
| 298 |
+
"loss": 7.3029,
|
| 299 |
+
"step": 380
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 8.12,
|
| 303 |
+
"learning_rate": 0.0003949645360130412,
|
| 304 |
+
"loss": 7.7765,
|
| 305 |
+
"step": 390
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 8.33,
|
| 309 |
+
"learning_rate": 0.0003946653194496012,
|
| 310 |
+
"loss": 7.2986,
|
| 311 |
+
"step": 400
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 8.33,
|
| 315 |
+
"eval_accuracy": 0.051572554180051966,
|
| 316 |
+
"eval_loss": 7.356107234954834,
|
| 317 |
+
"eval_runtime": 180.5938,
|
| 318 |
+
"eval_samples_per_second": 133.199,
|
| 319 |
+
"eval_steps_per_second": 4.164,
|
| 320 |
+
"step": 400
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 8.53,
|
| 324 |
+
"learning_rate": 0.00039435758759446025,
|
| 325 |
+
"loss": 7.3093,
|
| 326 |
+
"step": 410
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.74,
|
| 330 |
+
"learning_rate": 0.00039404135390880664,
|
| 331 |
+
"loss": 7.294,
|
| 332 |
+
"step": 420
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 8.95,
|
| 336 |
+
"learning_rate": 0.0003937166322257262,
|
| 337 |
+
"loss": 7.3083,
|
| 338 |
+
"step": 430
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 9.16,
|
| 342 |
+
"learning_rate": 0.00039338343674959745,
|
| 343 |
+
"loss": 7.7912,
|
| 344 |
+
"step": 440
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 9.37,
|
| 348 |
+
"learning_rate": 0.00039304178205546976,
|
| 349 |
+
"loss": 7.289,
|
| 350 |
+
"step": 450
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 9.37,
|
| 354 |
+
"eval_accuracy": 0.05145224079666028,
|
| 355 |
+
"eval_loss": 7.34950590133667,
|
| 356 |
+
"eval_runtime": 180.7201,
|
| 357 |
+
"eval_samples_per_second": 133.106,
|
| 358 |
+
"eval_steps_per_second": 4.161,
|
| 359 |
+
"step": 450
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 9.58,
|
| 363 |
+
"learning_rate": 0.00039269168308842634,
|
| 364 |
+
"loss": 7.3004,
|
| 365 |
+
"step": 460
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 9.78,
|
| 369 |
+
"learning_rate": 0.00039233315516293006,
|
| 370 |
+
"loss": 7.2938,
|
| 371 |
+
"step": 470
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 9.99,
|
| 375 |
+
"learning_rate": 0.00039196621396215403,
|
| 376 |
+
"loss": 7.2897,
|
| 377 |
+
"step": 480
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 10.21,
|
| 381 |
+
"learning_rate": 0.000391590875537295,
|
| 382 |
+
"loss": 7.7652,
|
| 383 |
+
"step": 490
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 10.41,
|
| 387 |
+
"learning_rate": 0.00039120715630687155,
|
| 388 |
+
"loss": 7.2879,
|
| 389 |
+
"step": 500
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 10.41,
|
| 393 |
+
"eval_accuracy": 0.05138556381472711,
|
| 394 |
+
"eval_loss": 7.3455071449279785,
|
| 395 |
+
"eval_runtime": 180.6339,
|
| 396 |
+
"eval_samples_per_second": 133.17,
|
| 397 |
+
"eval_steps_per_second": 4.163,
|
| 398 |
+
"step": 500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 10.62,
|
| 402 |
+
"learning_rate": 0.000390815073056006,
|
| 403 |
+
"loss": 7.2942,
|
| 404 |
+
"step": 510
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 10.82,
|
| 408 |
+
"learning_rate": 0.00039041464293568983,
|
| 409 |
+
"loss": 7.306,
|
| 410 |
+
"step": 520
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 11.04,
|
| 414 |
+
"learning_rate": 0.00039000588346203374,
|
| 415 |
+
"loss": 7.7754,
|
| 416 |
+
"step": 530
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 11.25,
|
| 420 |
+
"learning_rate": 0.0003895888125155014,
|
| 421 |
+
"loss": 7.2912,
|
| 422 |
+
"step": 540
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 11.45,
|
| 426 |
+
"learning_rate": 0.00038916344834012695,
|
| 427 |
+
"loss": 7.276,
|
| 428 |
+
"step": 550
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 11.45,
|
| 432 |
+
"eval_accuracy": 0.05130612004196204,
|
| 433 |
+
"eval_loss": 7.347738265991211,
|
| 434 |
+
"eval_runtime": 180.7636,
|
| 435 |
+
"eval_samples_per_second": 133.074,
|
| 436 |
+
"eval_steps_per_second": 4.16,
|
| 437 |
+
"step": 550
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 11.66,
|
| 441 |
+
"learning_rate": 0.00038872980954271757,
|
| 442 |
+
"loss": 7.3135,
|
| 443 |
+
"step": 560
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 11.86,
|
| 447 |
+
"learning_rate": 0.00038828791509203895,
|
| 448 |
+
"loss": 7.2859,
|
| 449 |
+
"step": 570
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 12.08,
|
| 453 |
+
"learning_rate": 0.00038783778431798597,
|
| 454 |
+
"loss": 7.7845,
|
| 455 |
+
"step": 580
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 12.29,
|
| 459 |
+
"learning_rate": 0.0003873794369107369,
|
| 460 |
+
"loss": 7.2966,
|
| 461 |
+
"step": 590
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 12.49,
|
| 465 |
+
"learning_rate": 0.0003869128929198922,
|
| 466 |
+
"loss": 7.3072,
|
| 467 |
+
"step": 600
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 12.49,
|
| 471 |
+
"eval_accuracy": 0.051627819878485845,
|
| 472 |
+
"eval_loss": 7.344621658325195,
|
| 473 |
+
"eval_runtime": 180.6519,
|
| 474 |
+
"eval_samples_per_second": 133.157,
|
| 475 |
+
"eval_steps_per_second": 4.163,
|
| 476 |
+
"step": 600
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 12.7,
|
| 480 |
+
"learning_rate": 0.0003864381727535973,
|
| 481 |
+
"loss": 7.3026,
|
| 482 |
+
"step": 610
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 12.9,
|
| 486 |
+
"learning_rate": 0.00038595529717765027,
|
| 487 |
+
"loss": 7.2966,
|
| 488 |
+
"step": 620
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 13.12,
|
| 492 |
+
"learning_rate": 0.0003854642873145931,
|
| 493 |
+
"loss": 7.7848,
|
| 494 |
+
"step": 630
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 13.33,
|
| 498 |
+
"learning_rate": 0.00038496516464278776,
|
| 499 |
+
"loss": 7.2964,
|
| 500 |
+
"step": 640
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 13.53,
|
| 504 |
+
"learning_rate": 0.00038445795099547697,
|
| 505 |
+
"loss": 7.2978,
|
| 506 |
+
"step": 650
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 13.53,
|
| 510 |
+
"eval_accuracy": 0.05143096217098587,
|
| 511 |
+
"eval_loss": 7.346319198608398,
|
| 512 |
+
"eval_runtime": 180.763,
|
| 513 |
+
"eval_samples_per_second": 133.075,
|
| 514 |
+
"eval_steps_per_second": 4.16,
|
| 515 |
+
"step": 650
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 13.74,
|
| 519 |
+
"learning_rate": 0.0003839426685598287,
|
| 520 |
+
"loss": 7.2919,
|
| 521 |
+
"step": 660
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 13.95,
|
| 525 |
+
"learning_rate": 0.000383419339875966,
|
| 526 |
+
"loss": 7.3006,
|
| 527 |
+
"step": 670
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 14.16,
|
| 531 |
+
"learning_rate": 0.00038288798783598087,
|
| 532 |
+
"loss": 7.7738,
|
| 533 |
+
"step": 680
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 14.37,
|
| 537 |
+
"learning_rate": 0.0003823486356829329,
|
| 538 |
+
"loss": 7.2839,
|
| 539 |
+
"step": 690
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 14.58,
|
| 543 |
+
"learning_rate": 0.0003818013070098325,
|
| 544 |
+
"loss": 7.2857,
|
| 545 |
+
"step": 700
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 14.58,
|
| 549 |
+
"eval_accuracy": 0.05146984844436126,
|
| 550 |
+
"eval_loss": 7.342555999755859,
|
| 551 |
+
"eval_runtime": 180.8063,
|
| 552 |
+
"eval_samples_per_second": 133.043,
|
| 553 |
+
"eval_steps_per_second": 4.159,
|
| 554 |
+
"step": 700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 14.78,
|
| 558 |
+
"learning_rate": 0.0003812460257586089,
|
| 559 |
+
"loss": 7.2949,
|
| 560 |
+
"step": 710
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 14.99,
|
| 564 |
+
"learning_rate": 0.000380682816219063,
|
| 565 |
+
"loss": 7.3249,
|
| 566 |
+
"step": 720
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 15.21,
|
| 570 |
+
"learning_rate": 0.00038011170302780446,
|
| 571 |
+
"loss": 7.7486,
|
| 572 |
+
"step": 730
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 15.41,
|
| 576 |
+
"learning_rate": 0.00037953271116717444,
|
| 577 |
+
"loss": 7.2879,
|
| 578 |
+
"step": 740
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 15.62,
|
| 582 |
+
"learning_rate": 0.0003789458659641527,
|
| 583 |
+
"loss": 7.2868,
|
| 584 |
+
"step": 750
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 15.62,
|
| 588 |
+
"eval_accuracy": 0.05147383671825258,
|
| 589 |
+
"eval_loss": 7.343778610229492,
|
| 590 |
+
"eval_runtime": 180.8254,
|
| 591 |
+
"eval_samples_per_second": 133.029,
|
| 592 |
+
"eval_steps_per_second": 4.159,
|
| 593 |
+
"step": 750
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 15.82,
|
| 597 |
+
"learning_rate": 0.0003783511930892495,
|
| 598 |
+
"loss": 7.2986,
|
| 599 |
+
"step": 760
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 16.04,
|
| 603 |
+
"learning_rate": 0.00037774871855538275,
|
| 604 |
+
"loss": 7.7788,
|
| 605 |
+
"step": 770
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 16.25,
|
| 609 |
+
"learning_rate": 0.00037713846871674045,
|
| 610 |
+
"loss": 7.2858,
|
| 611 |
+
"step": 780
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 16.45,
|
| 615 |
+
"learning_rate": 0.0003765204702676274,
|
| 616 |
+
"loss": 7.2937,
|
| 617 |
+
"step": 790
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 16.66,
|
| 621 |
+
"learning_rate": 0.0003758947502412978,
|
| 622 |
+
"loss": 7.2973,
|
| 623 |
+
"step": 800
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 16.66,
|
| 627 |
+
"eval_accuracy": 0.051658592501666364,
|
| 628 |
+
"eval_loss": 7.344185829162598,
|
| 629 |
+
"eval_runtime": 180.7375,
|
| 630 |
+
"eval_samples_per_second": 133.094,
|
| 631 |
+
"eval_steps_per_second": 4.161,
|
| 632 |
+
"step": 800
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 16.86,
|
| 636 |
+
"learning_rate": 0.0003752613360087727,
|
| 637 |
+
"loss": 7.3043,
|
| 638 |
+
"step": 810
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 17.08,
|
| 642 |
+
"learning_rate": 0.00037462025527764265,
|
| 643 |
+
"loss": 7.7616,
|
| 644 |
+
"step": 820
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 17.29,
|
| 648 |
+
"learning_rate": 0.00037397153609085553,
|
| 649 |
+
"loss": 7.2869,
|
| 650 |
+
"step": 830
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 17.49,
|
| 654 |
+
"learning_rate": 0.0003733152068254901,
|
| 655 |
+
"loss": 7.2798,
|
| 656 |
+
"step": 840
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 17.7,
|
| 660 |
+
"learning_rate": 0.00037265129619151483,
|
| 661 |
+
"loss": 7.2988,
|
| 662 |
+
"step": 850
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 17.7,
|
| 666 |
+
"eval_accuracy": 0.051239018394020945,
|
| 667 |
+
"eval_loss": 7.343734264373779,
|
| 668 |
+
"eval_runtime": 180.5675,
|
| 669 |
+
"eval_samples_per_second": 133.219,
|
| 670 |
+
"eval_steps_per_second": 4.165,
|
| 671 |
+
"step": 850
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 17.7,
|
| 675 |
+
"step": 850,
|
| 676 |
+
"total_flos": 1.1524171581514752e+17,
|
| 677 |
+
"train_loss": 7.482659651812385,
|
| 678 |
+
"train_runtime": 11122.8848,
|
| 679 |
+
"train_samples_per_second": 223.953,
|
| 680 |
+
"train_steps_per_second": 0.432
|
| 681 |
+
}
|
| 682 |
+
],
|
| 683 |
+
"max_steps": 4800,
|
| 684 |
+
"num_train_epochs": 100,
|
| 685 |
+
"total_flos": 1.1524171581514752e+17,
|
| 686 |
+
"trial_name": null,
|
| 687 |
+
"trial_params": null
|
| 688 |
+
}
|