Commit
•
9de37e7
1
Parent(s):
5640ecd
Upload trainer_state.json
Browse files- trainer_state.json +184 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,184 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 11120,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.18,
|
| 12 |
+
"learning_rate": 4.8501199040767384e-05,
|
| 13 |
+
"loss": 2.0957,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.36,
|
| 18 |
+
"learning_rate": 4.700239808153478e-05,
|
| 19 |
+
"loss": 2.022,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.54,
|
| 24 |
+
"learning_rate": 4.550359712230216e-05,
|
| 25 |
+
"loss": 1.9842,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.72,
|
| 30 |
+
"learning_rate": 4.400479616306955e-05,
|
| 31 |
+
"loss": 1.9624,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.9,
|
| 36 |
+
"learning_rate": 4.2505995203836936e-05,
|
| 37 |
+
"loss": 1.9237,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0,
|
| 42 |
+
"eval_bleu": 25.317,
|
| 43 |
+
"eval_gen_len": 11.9956,
|
| 44 |
+
"eval_loss": 1.8112746477127075,
|
| 45 |
+
"eval_runtime": 164.477,
|
| 46 |
+
"eval_samples_per_second": 12.518,
|
| 47 |
+
"step": 2780
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.08,
|
| 51 |
+
"learning_rate": 4.100719424460432e-05,
|
| 52 |
+
"loss": 1.7808,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.26,
|
| 57 |
+
"learning_rate": 3.9508393285371705e-05,
|
| 58 |
+
"loss": 1.6002,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.44,
|
| 63 |
+
"learning_rate": 3.8009592326139094e-05,
|
| 64 |
+
"loss": 1.6191,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.62,
|
| 69 |
+
"learning_rate": 3.6510791366906475e-05,
|
| 70 |
+
"loss": 1.6215,
|
| 71 |
+
"step": 4500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.8,
|
| 75 |
+
"learning_rate": 3.501199040767386e-05,
|
| 76 |
+
"loss": 1.6134,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.98,
|
| 81 |
+
"learning_rate": 3.351318944844125e-05,
|
| 82 |
+
"loss": 1.6013,
|
| 83 |
+
"step": 5500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.0,
|
| 87 |
+
"eval_bleu": 25.523,
|
| 88 |
+
"eval_gen_len": 12.0345,
|
| 89 |
+
"eval_loss": 1.8125885725021362,
|
| 90 |
+
"eval_runtime": 182.336,
|
| 91 |
+
"eval_samples_per_second": 11.292,
|
| 92 |
+
"step": 5560
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.16,
|
| 96 |
+
"learning_rate": 3.201438848920863e-05,
|
| 97 |
+
"loss": 1.3405,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.34,
|
| 102 |
+
"learning_rate": 3.051558752997602e-05,
|
| 103 |
+
"loss": 1.3232,
|
| 104 |
+
"step": 6500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.52,
|
| 108 |
+
"learning_rate": 2.9016786570743405e-05,
|
| 109 |
+
"loss": 1.3278,
|
| 110 |
+
"step": 7000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.7,
|
| 114 |
+
"learning_rate": 2.7517985611510793e-05,
|
| 115 |
+
"loss": 1.3295,
|
| 116 |
+
"step": 7500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.88,
|
| 120 |
+
"learning_rate": 2.6019184652278178e-05,
|
| 121 |
+
"loss": 1.335,
|
| 122 |
+
"step": 8000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 3.0,
|
| 126 |
+
"eval_bleu": 25.7471,
|
| 127 |
+
"eval_gen_len": 11.8786,
|
| 128 |
+
"eval_loss": 1.90364408493042,
|
| 129 |
+
"eval_runtime": 179.3491,
|
| 130 |
+
"eval_samples_per_second": 11.48,
|
| 131 |
+
"step": 8340
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.06,
|
| 135 |
+
"learning_rate": 2.4520383693045566e-05,
|
| 136 |
+
"loss": 1.2511,
|
| 137 |
+
"step": 8500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 3.24,
|
| 141 |
+
"learning_rate": 2.302158273381295e-05,
|
| 142 |
+
"loss": 1.0838,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.42,
|
| 147 |
+
"learning_rate": 2.1522781774580335e-05,
|
| 148 |
+
"loss": 1.0898,
|
| 149 |
+
"step": 9500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 3.6,
|
| 153 |
+
"learning_rate": 2.0023980815347724e-05,
|
| 154 |
+
"loss": 1.0904,
|
| 155 |
+
"step": 10000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.78,
|
| 159 |
+
"learning_rate": 1.8525179856115108e-05,
|
| 160 |
+
"loss": 1.0924,
|
| 161 |
+
"step": 10500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 3.96,
|
| 165 |
+
"learning_rate": 1.7026378896882493e-05,
|
| 166 |
+
"loss": 1.0952,
|
| 167 |
+
"step": 11000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.0,
|
| 171 |
+
"eval_bleu": 24.7279,
|
| 172 |
+
"eval_gen_len": 12.0661,
|
| 173 |
+
"eval_loss": 2.009955406188965,
|
| 174 |
+
"eval_runtime": 180.8445,
|
| 175 |
+
"eval_samples_per_second": 11.385,
|
| 176 |
+
"step": 11120
|
| 177 |
+
}
|
| 178 |
+
],
|
| 179 |
+
"max_steps": 16680,
|
| 180 |
+
"num_train_epochs": 6,
|
| 181 |
+
"total_flos": 1.2244523192274125e+17,
|
| 182 |
+
"trial_name": null,
|
| 183 |
+
"trial_params": null
|
| 184 |
+
}
|