longt5_xl_gov_memsum_bp_20 / trainer_state.json
learn3r's picture
End of training
99c281a
{
"best_metric": 3.6259207725524902,
"best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_gov_memsum_bp_20/checkpoint-136",
"epoch": 4.985679917516325,
"eval_steps": 500,
"global_step": 680,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.001,
"loss": 0.1745,
"step": 2
},
{
"epoch": 0.03,
"learning_rate": 0.001,
"loss": 0.1678,
"step": 4
},
{
"epoch": 0.04,
"learning_rate": 0.001,
"loss": 0.1686,
"step": 6
},
{
"epoch": 0.06,
"learning_rate": 0.001,
"loss": 0.1572,
"step": 8
},
{
"epoch": 0.07,
"learning_rate": 0.001,
"loss": 0.1619,
"step": 10
},
{
"epoch": 0.09,
"learning_rate": 0.001,
"loss": 0.1668,
"step": 12
},
{
"epoch": 0.1,
"learning_rate": 0.001,
"loss": 0.1699,
"step": 14
},
{
"epoch": 0.12,
"learning_rate": 0.001,
"loss": 0.1651,
"step": 16
},
{
"epoch": 0.13,
"learning_rate": 0.001,
"loss": 0.1616,
"step": 18
},
{
"epoch": 0.15,
"learning_rate": 0.001,
"loss": 0.1565,
"step": 20
},
{
"epoch": 0.16,
"learning_rate": 0.001,
"loss": 0.1558,
"step": 22
},
{
"epoch": 0.18,
"learning_rate": 0.001,
"loss": 0.1532,
"step": 24
},
{
"epoch": 0.19,
"learning_rate": 0.001,
"loss": 0.1623,
"step": 26
},
{
"epoch": 0.21,
"learning_rate": 0.001,
"loss": 0.1571,
"step": 28
},
{
"epoch": 0.22,
"learning_rate": 0.001,
"loss": 0.1532,
"step": 30
},
{
"epoch": 0.23,
"learning_rate": 0.001,
"loss": 0.164,
"step": 32
},
{
"epoch": 0.25,
"learning_rate": 0.001,
"loss": 0.1516,
"step": 34
},
{
"epoch": 0.26,
"learning_rate": 0.001,
"loss": 0.1572,
"step": 36
},
{
"epoch": 0.28,
"learning_rate": 0.001,
"loss": 0.1609,
"step": 38
},
{
"epoch": 0.29,
"learning_rate": 0.001,
"loss": 0.1581,
"step": 40
},
{
"epoch": 0.31,
"learning_rate": 0.001,
"loss": 0.1586,
"step": 42
},
{
"epoch": 0.32,
"learning_rate": 0.001,
"loss": 0.1621,
"step": 44
},
{
"epoch": 0.34,
"learning_rate": 0.001,
"loss": 0.1573,
"step": 46
},
{
"epoch": 0.35,
"learning_rate": 0.001,
"loss": 0.1541,
"step": 48
},
{
"epoch": 0.37,
"learning_rate": 0.001,
"loss": 0.1584,
"step": 50
},
{
"epoch": 0.38,
"learning_rate": 0.001,
"loss": 0.1574,
"step": 52
},
{
"epoch": 0.4,
"learning_rate": 0.001,
"loss": 0.1505,
"step": 54
},
{
"epoch": 0.41,
"learning_rate": 0.001,
"loss": 0.1567,
"step": 56
},
{
"epoch": 0.43,
"learning_rate": 0.001,
"loss": 0.1573,
"step": 58
},
{
"epoch": 0.44,
"learning_rate": 0.001,
"loss": 0.1584,
"step": 60
},
{
"epoch": 0.45,
"learning_rate": 0.001,
"loss": 0.1561,
"step": 62
},
{
"epoch": 0.47,
"learning_rate": 0.001,
"loss": 0.1543,
"step": 64
},
{
"epoch": 0.48,
"learning_rate": 0.001,
"loss": 0.1539,
"step": 66
},
{
"epoch": 0.5,
"learning_rate": 0.001,
"loss": 0.152,
"step": 68
},
{
"epoch": 0.51,
"learning_rate": 0.001,
"loss": 0.1485,
"step": 70
},
{
"epoch": 0.53,
"learning_rate": 0.001,
"loss": 0.1533,
"step": 72
},
{
"epoch": 0.54,
"learning_rate": 0.001,
"loss": 0.1653,
"step": 74
},
{
"epoch": 0.56,
"learning_rate": 0.001,
"loss": 0.1607,
"step": 76
},
{
"epoch": 0.57,
"learning_rate": 0.001,
"loss": 0.1625,
"step": 78
},
{
"epoch": 0.59,
"learning_rate": 0.001,
"loss": 0.1507,
"step": 80
},
{
"epoch": 0.6,
"learning_rate": 0.001,
"loss": 0.1605,
"step": 82
},
{
"epoch": 0.62,
"learning_rate": 0.001,
"loss": 0.1579,
"step": 84
},
{
"epoch": 0.63,
"learning_rate": 0.001,
"loss": 0.1525,
"step": 86
},
{
"epoch": 0.65,
"learning_rate": 0.001,
"loss": 0.156,
"step": 88
},
{
"epoch": 0.66,
"learning_rate": 0.001,
"loss": 0.1605,
"step": 90
},
{
"epoch": 0.67,
"learning_rate": 0.001,
"loss": 0.1586,
"step": 92
},
{
"epoch": 0.69,
"learning_rate": 0.001,
"loss": 0.1548,
"step": 94
},
{
"epoch": 0.7,
"learning_rate": 0.001,
"loss": 0.1491,
"step": 96
},
{
"epoch": 0.72,
"learning_rate": 0.001,
"loss": 0.1489,
"step": 98
},
{
"epoch": 0.73,
"learning_rate": 0.001,
"loss": 0.1596,
"step": 100
},
{
"epoch": 0.75,
"learning_rate": 0.001,
"loss": 0.1615,
"step": 102
},
{
"epoch": 0.76,
"learning_rate": 0.001,
"loss": 0.1638,
"step": 104
},
{
"epoch": 0.78,
"learning_rate": 0.001,
"loss": 0.1577,
"step": 106
},
{
"epoch": 0.79,
"learning_rate": 0.001,
"loss": 0.1552,
"step": 108
},
{
"epoch": 0.81,
"learning_rate": 0.001,
"loss": 0.1592,
"step": 110
},
{
"epoch": 0.82,
"learning_rate": 0.001,
"loss": 0.1632,
"step": 112
},
{
"epoch": 0.84,
"learning_rate": 0.001,
"loss": 0.1548,
"step": 114
},
{
"epoch": 0.85,
"learning_rate": 0.001,
"loss": 0.1509,
"step": 116
},
{
"epoch": 0.87,
"learning_rate": 0.001,
"loss": 0.1539,
"step": 118
},
{
"epoch": 0.88,
"learning_rate": 0.001,
"loss": 0.1584,
"step": 120
},
{
"epoch": 0.89,
"learning_rate": 0.001,
"loss": 0.1663,
"step": 122
},
{
"epoch": 0.91,
"learning_rate": 0.001,
"loss": 0.1628,
"step": 124
},
{
"epoch": 0.92,
"learning_rate": 0.001,
"loss": 0.1529,
"step": 126
},
{
"epoch": 0.94,
"learning_rate": 0.001,
"loss": 0.1506,
"step": 128
},
{
"epoch": 0.95,
"learning_rate": 0.001,
"loss": 0.1595,
"step": 130
},
{
"epoch": 0.97,
"learning_rate": 0.001,
"loss": 0.1564,
"step": 132
},
{
"epoch": 0.98,
"learning_rate": 0.001,
"loss": 0.1561,
"step": 134
},
{
"epoch": 1.0,
"learning_rate": 0.001,
"loss": 0.1591,
"step": 136
},
{
"epoch": 1.0,
"eval_gen_len": 1510.8694758478932,
"eval_loss": 3.6259207725524902,
"eval_rouge1": 42.5601,
"eval_rouge2": 14.1791,
"eval_rougeL": 17.9691,
"eval_rougeLsum": 40.487,
"eval_runtime": 18906.076,
"eval_samples_per_second": 0.051,
"eval_steps_per_second": 0.006,
"step": 136
},
{
"epoch": 1.01,
"learning_rate": 0.001,
"loss": 0.1101,
"step": 138
},
{
"epoch": 1.03,
"learning_rate": 0.001,
"loss": 0.099,
"step": 140
},
{
"epoch": 1.04,
"learning_rate": 0.001,
"loss": 0.0929,
"step": 142
},
{
"epoch": 1.06,
"learning_rate": 0.001,
"loss": 0.0896,
"step": 144
},
{
"epoch": 1.07,
"learning_rate": 0.001,
"loss": 0.0955,
"step": 146
},
{
"epoch": 1.09,
"learning_rate": 0.001,
"loss": 0.094,
"step": 148
},
{
"epoch": 1.1,
"learning_rate": 0.001,
"loss": 0.096,
"step": 150
},
{
"epoch": 1.11,
"learning_rate": 0.001,
"loss": 0.0938,
"step": 152
},
{
"epoch": 1.13,
"learning_rate": 0.001,
"loss": 0.0959,
"step": 154
},
{
"epoch": 1.14,
"learning_rate": 0.001,
"loss": 0.0987,
"step": 156
},
{
"epoch": 1.16,
"learning_rate": 0.001,
"loss": 0.1012,
"step": 158
},
{
"epoch": 1.17,
"learning_rate": 0.001,
"loss": 0.1017,
"step": 160
},
{
"epoch": 1.19,
"learning_rate": 0.001,
"loss": 0.1003,
"step": 162
},
{
"epoch": 1.2,
"learning_rate": 0.001,
"loss": 0.1055,
"step": 164
},
{
"epoch": 1.22,
"learning_rate": 0.001,
"loss": 0.105,
"step": 166
},
{
"epoch": 1.23,
"learning_rate": 0.001,
"loss": 0.1026,
"step": 168
},
{
"epoch": 1.25,
"learning_rate": 0.001,
"loss": 0.0987,
"step": 170
},
{
"epoch": 1.26,
"learning_rate": 0.001,
"loss": 0.1018,
"step": 172
},
{
"epoch": 1.28,
"learning_rate": 0.001,
"loss": 0.1062,
"step": 174
},
{
"epoch": 1.29,
"learning_rate": 0.001,
"loss": 0.1055,
"step": 176
},
{
"epoch": 1.31,
"learning_rate": 0.001,
"loss": 0.1005,
"step": 178
},
{
"epoch": 1.32,
"learning_rate": 0.001,
"loss": 0.1047,
"step": 180
},
{
"epoch": 1.33,
"learning_rate": 0.001,
"loss": 0.1065,
"step": 182
},
{
"epoch": 1.35,
"learning_rate": 0.001,
"loss": 0.1082,
"step": 184
},
{
"epoch": 1.36,
"learning_rate": 0.001,
"loss": 0.1068,
"step": 186
},
{
"epoch": 1.38,
"learning_rate": 0.001,
"loss": 0.1074,
"step": 188
},
{
"epoch": 1.39,
"learning_rate": 0.001,
"loss": 0.1077,
"step": 190
},
{
"epoch": 1.41,
"learning_rate": 0.001,
"loss": 0.1109,
"step": 192
},
{
"epoch": 1.42,
"learning_rate": 0.001,
"loss": 0.1011,
"step": 194
},
{
"epoch": 1.44,
"learning_rate": 0.001,
"loss": 0.1015,
"step": 196
},
{
"epoch": 1.45,
"learning_rate": 0.001,
"loss": 0.1015,
"step": 198
},
{
"epoch": 1.47,
"learning_rate": 0.001,
"loss": 0.1094,
"step": 200
},
{
"epoch": 1.48,
"learning_rate": 0.001,
"loss": 0.108,
"step": 202
},
{
"epoch": 1.5,
"learning_rate": 0.001,
"loss": 0.1041,
"step": 204
},
{
"epoch": 1.51,
"learning_rate": 0.001,
"loss": 0.1092,
"step": 206
},
{
"epoch": 1.53,
"learning_rate": 0.001,
"loss": 0.1085,
"step": 208
},
{
"epoch": 1.54,
"learning_rate": 0.001,
"loss": 0.1079,
"step": 210
},
{
"epoch": 1.55,
"learning_rate": 0.001,
"loss": 0.1074,
"step": 212
},
{
"epoch": 1.57,
"learning_rate": 0.001,
"loss": 0.1111,
"step": 214
},
{
"epoch": 1.58,
"learning_rate": 0.001,
"loss": 0.1153,
"step": 216
},
{
"epoch": 1.6,
"learning_rate": 0.001,
"loss": 0.1127,
"step": 218
},
{
"epoch": 1.61,
"learning_rate": 0.001,
"loss": 0.1141,
"step": 220
},
{
"epoch": 1.63,
"learning_rate": 0.001,
"loss": 0.1196,
"step": 222
},
{
"epoch": 1.64,
"learning_rate": 0.001,
"loss": 0.1138,
"step": 224
},
{
"epoch": 1.66,
"learning_rate": 0.001,
"loss": 0.1136,
"step": 226
},
{
"epoch": 1.67,
"learning_rate": 0.001,
"loss": 0.1147,
"step": 228
},
{
"epoch": 1.69,
"learning_rate": 0.001,
"loss": 0.1091,
"step": 230
},
{
"epoch": 1.7,
"learning_rate": 0.001,
"loss": 0.1104,
"step": 232
},
{
"epoch": 1.72,
"learning_rate": 0.001,
"loss": 0.1164,
"step": 234
},
{
"epoch": 1.73,
"learning_rate": 0.001,
"loss": 0.112,
"step": 236
},
{
"epoch": 1.74,
"learning_rate": 0.001,
"loss": 0.1132,
"step": 238
},
{
"epoch": 1.76,
"learning_rate": 0.001,
"loss": 0.1088,
"step": 240
},
{
"epoch": 1.77,
"learning_rate": 0.001,
"loss": 0.1126,
"step": 242
},
{
"epoch": 1.79,
"learning_rate": 0.001,
"loss": 0.112,
"step": 244
},
{
"epoch": 1.8,
"learning_rate": 0.001,
"loss": 0.1112,
"step": 246
},
{
"epoch": 1.82,
"learning_rate": 0.001,
"loss": 0.117,
"step": 248
},
{
"epoch": 1.83,
"learning_rate": 0.001,
"loss": 0.1128,
"step": 250
},
{
"epoch": 1.85,
"learning_rate": 0.001,
"loss": 0.1087,
"step": 252
},
{
"epoch": 1.86,
"learning_rate": 0.001,
"loss": 0.1149,
"step": 254
},
{
"epoch": 1.88,
"learning_rate": 0.001,
"loss": 0.1164,
"step": 256
},
{
"epoch": 1.89,
"learning_rate": 0.001,
"loss": 0.1199,
"step": 258
},
{
"epoch": 1.91,
"learning_rate": 0.001,
"loss": 0.1161,
"step": 260
},
{
"epoch": 1.92,
"learning_rate": 0.001,
"loss": 0.1136,
"step": 262
},
{
"epoch": 1.94,
"learning_rate": 0.001,
"loss": 0.111,
"step": 264
},
{
"epoch": 1.95,
"learning_rate": 0.001,
"loss": 0.1141,
"step": 266
},
{
"epoch": 1.96,
"learning_rate": 0.001,
"loss": 0.1109,
"step": 268
},
{
"epoch": 1.98,
"learning_rate": 0.001,
"loss": 0.1133,
"step": 270
},
{
"epoch": 1.99,
"learning_rate": 0.001,
"loss": 0.1186,
"step": 272
},
{
"epoch": 1.99,
"eval_gen_len": 1707.0400822199383,
"eval_loss": 3.788456678390503,
"eval_rouge1": 39.795,
"eval_rouge2": 13.2493,
"eval_rougeL": 17.3095,
"eval_rougeLsum": 37.9065,
"eval_runtime": 19648.4,
"eval_samples_per_second": 0.05,
"eval_steps_per_second": 0.006,
"step": 272
},
{
"epoch": 2.01,
"learning_rate": 0.001,
"loss": 0.0991,
"step": 274
},
{
"epoch": 2.02,
"learning_rate": 0.001,
"loss": 0.0755,
"step": 276
},
{
"epoch": 2.04,
"learning_rate": 0.001,
"loss": 0.0746,
"step": 278
},
{
"epoch": 2.05,
"learning_rate": 0.001,
"loss": 0.0766,
"step": 280
},
{
"epoch": 2.07,
"learning_rate": 0.001,
"loss": 0.0757,
"step": 282
},
{
"epoch": 2.08,
"learning_rate": 0.001,
"loss": 0.0772,
"step": 284
},
{
"epoch": 2.1,
"learning_rate": 0.001,
"loss": 0.0778,
"step": 286
},
{
"epoch": 2.11,
"learning_rate": 0.001,
"loss": 0.0788,
"step": 288
},
{
"epoch": 2.13,
"learning_rate": 0.001,
"loss": 0.0763,
"step": 290
},
{
"epoch": 2.14,
"learning_rate": 0.001,
"loss": 0.0826,
"step": 292
},
{
"epoch": 2.16,
"learning_rate": 0.001,
"loss": 0.0814,
"step": 294
},
{
"epoch": 2.17,
"learning_rate": 0.001,
"loss": 0.08,
"step": 296
},
{
"epoch": 2.18,
"learning_rate": 0.001,
"loss": 0.0799,
"step": 298
},
{
"epoch": 2.2,
"learning_rate": 0.001,
"loss": 0.0826,
"step": 300
},
{
"epoch": 2.21,
"learning_rate": 0.001,
"loss": 0.0791,
"step": 302
},
{
"epoch": 2.23,
"learning_rate": 0.001,
"loss": 0.0821,
"step": 304
},
{
"epoch": 2.24,
"learning_rate": 0.001,
"loss": 0.0846,
"step": 306
},
{
"epoch": 2.26,
"learning_rate": 0.001,
"loss": 0.081,
"step": 308
},
{
"epoch": 2.27,
"learning_rate": 0.001,
"loss": 0.0841,
"step": 310
},
{
"epoch": 2.29,
"learning_rate": 0.001,
"loss": 0.0838,
"step": 312
},
{
"epoch": 2.3,
"learning_rate": 0.001,
"loss": 0.0875,
"step": 314
},
{
"epoch": 2.32,
"learning_rate": 0.001,
"loss": 0.0871,
"step": 316
},
{
"epoch": 2.33,
"learning_rate": 0.001,
"loss": 0.0869,
"step": 318
},
{
"epoch": 2.35,
"learning_rate": 0.001,
"loss": 0.0905,
"step": 320
},
{
"epoch": 2.36,
"learning_rate": 0.001,
"loss": 0.0947,
"step": 322
},
{
"epoch": 2.38,
"learning_rate": 0.001,
"loss": 0.1041,
"step": 324
},
{
"epoch": 2.39,
"learning_rate": 0.001,
"loss": 0.0915,
"step": 326
},
{
"epoch": 2.4,
"learning_rate": 0.001,
"loss": 0.0869,
"step": 328
},
{
"epoch": 2.42,
"learning_rate": 0.001,
"loss": 0.0854,
"step": 330
},
{
"epoch": 2.43,
"learning_rate": 0.001,
"loss": 0.0925,
"step": 332
},
{
"epoch": 2.45,
"learning_rate": 0.001,
"loss": 0.0961,
"step": 334
},
{
"epoch": 2.46,
"learning_rate": 0.001,
"loss": 0.0922,
"step": 336
},
{
"epoch": 2.48,
"learning_rate": 0.001,
"loss": 0.0847,
"step": 338
},
{
"epoch": 2.49,
"learning_rate": 0.001,
"loss": 0.0906,
"step": 340
},
{
"epoch": 2.51,
"learning_rate": 0.001,
"loss": 0.0891,
"step": 342
},
{
"epoch": 2.52,
"learning_rate": 0.001,
"loss": 0.0916,
"step": 344
},
{
"epoch": 2.54,
"learning_rate": 0.001,
"loss": 0.0907,
"step": 346
},
{
"epoch": 2.55,
"learning_rate": 0.001,
"loss": 0.0912,
"step": 348
},
{
"epoch": 2.57,
"learning_rate": 0.001,
"loss": 0.0857,
"step": 350
},
{
"epoch": 2.58,
"learning_rate": 0.001,
"loss": 0.0864,
"step": 352
},
{
"epoch": 2.6,
"learning_rate": 0.001,
"loss": 0.087,
"step": 354
},
{
"epoch": 2.61,
"learning_rate": 0.001,
"loss": 0.0957,
"step": 356
},
{
"epoch": 2.62,
"learning_rate": 0.001,
"loss": 0.089,
"step": 358
},
{
"epoch": 2.64,
"learning_rate": 0.001,
"loss": 0.0939,
"step": 360
},
{
"epoch": 2.65,
"learning_rate": 0.001,
"loss": 0.0891,
"step": 362
},
{
"epoch": 2.67,
"learning_rate": 0.001,
"loss": 0.0912,
"step": 364
},
{
"epoch": 2.68,
"learning_rate": 0.001,
"loss": 0.0932,
"step": 366
},
{
"epoch": 2.7,
"learning_rate": 0.001,
"loss": 0.0932,
"step": 368
},
{
"epoch": 2.71,
"learning_rate": 0.001,
"loss": 0.0923,
"step": 370
},
{
"epoch": 2.73,
"learning_rate": 0.001,
"loss": 0.0933,
"step": 372
},
{
"epoch": 2.74,
"learning_rate": 0.001,
"loss": 0.096,
"step": 374
},
{
"epoch": 2.76,
"learning_rate": 0.001,
"loss": 0.0923,
"step": 376
},
{
"epoch": 2.77,
"learning_rate": 0.001,
"loss": 0.0918,
"step": 378
},
{
"epoch": 2.79,
"learning_rate": 0.001,
"loss": 0.091,
"step": 380
},
{
"epoch": 2.8,
"learning_rate": 0.001,
"loss": 0.09,
"step": 382
},
{
"epoch": 2.82,
"learning_rate": 0.001,
"loss": 0.0894,
"step": 384
},
{
"epoch": 2.83,
"learning_rate": 0.001,
"loss": 0.0962,
"step": 386
},
{
"epoch": 2.84,
"learning_rate": 0.001,
"loss": 0.0923,
"step": 388
},
{
"epoch": 2.86,
"learning_rate": 0.001,
"loss": 0.0942,
"step": 390
},
{
"epoch": 2.87,
"learning_rate": 0.001,
"loss": 0.0999,
"step": 392
},
{
"epoch": 2.89,
"learning_rate": 0.001,
"loss": 0.0972,
"step": 394
},
{
"epoch": 2.9,
"learning_rate": 0.001,
"loss": 0.1022,
"step": 396
},
{
"epoch": 2.92,
"learning_rate": 0.001,
"loss": 0.0962,
"step": 398
},
{
"epoch": 2.93,
"learning_rate": 0.001,
"loss": 0.0917,
"step": 400
},
{
"epoch": 2.95,
"learning_rate": 0.001,
"loss": 0.0978,
"step": 402
},
{
"epoch": 2.96,
"learning_rate": 0.001,
"loss": 0.0909,
"step": 404
},
{
"epoch": 2.98,
"learning_rate": 0.001,
"loss": 0.0948,
"step": 406
},
{
"epoch": 2.99,
"learning_rate": 0.001,
"loss": 0.097,
"step": 408
},
{
"epoch": 3.0,
"eval_gen_len": 1442.7728674203495,
"eval_loss": 4.019211769104004,
"eval_rouge1": 41.4441,
"eval_rouge2": 13.4026,
"eval_rougeL": 17.8804,
"eval_rougeLsum": 39.4502,
"eval_runtime": 18768.077,
"eval_samples_per_second": 0.052,
"eval_steps_per_second": 0.007,
"step": 409
},
{
"epoch": 3.01,
"learning_rate": 0.001,
"loss": 0.0843,
"step": 410
},
{
"epoch": 3.02,
"learning_rate": 0.001,
"loss": 0.0682,
"step": 412
},
{
"epoch": 3.04,
"learning_rate": 0.001,
"loss": 0.0673,
"step": 414
},
{
"epoch": 3.05,
"learning_rate": 0.001,
"loss": 0.0644,
"step": 416
},
{
"epoch": 3.06,
"learning_rate": 0.001,
"loss": 0.0659,
"step": 418
},
{
"epoch": 3.08,
"learning_rate": 0.001,
"loss": 0.0675,
"step": 420
},
{
"epoch": 3.09,
"learning_rate": 0.001,
"loss": 0.0648,
"step": 422
},
{
"epoch": 3.11,
"learning_rate": 0.001,
"loss": 0.0664,
"step": 424
},
{
"epoch": 3.12,
"learning_rate": 0.001,
"loss": 0.0714,
"step": 426
},
{
"epoch": 3.14,
"learning_rate": 0.001,
"loss": 0.0703,
"step": 428
},
{
"epoch": 3.15,
"learning_rate": 0.001,
"loss": 0.068,
"step": 430
},
{
"epoch": 3.17,
"learning_rate": 0.001,
"loss": 0.0683,
"step": 432
},
{
"epoch": 3.18,
"learning_rate": 0.001,
"loss": 0.0676,
"step": 434
},
{
"epoch": 3.2,
"learning_rate": 0.001,
"loss": 0.0703,
"step": 436
},
{
"epoch": 3.21,
"learning_rate": 0.001,
"loss": 0.0726,
"step": 438
},
{
"epoch": 3.23,
"learning_rate": 0.001,
"loss": 0.0712,
"step": 440
},
{
"epoch": 3.24,
"learning_rate": 0.001,
"loss": 0.0706,
"step": 442
},
{
"epoch": 3.26,
"learning_rate": 0.001,
"loss": 0.0736,
"step": 444
},
{
"epoch": 3.27,
"learning_rate": 0.001,
"loss": 0.0698,
"step": 446
},
{
"epoch": 3.28,
"learning_rate": 0.001,
"loss": 0.0753,
"step": 448
},
{
"epoch": 3.3,
"learning_rate": 0.001,
"loss": 0.0768,
"step": 450
},
{
"epoch": 3.31,
"learning_rate": 0.001,
"loss": 0.0764,
"step": 452
},
{
"epoch": 3.33,
"learning_rate": 0.001,
"loss": 0.0748,
"step": 454
},
{
"epoch": 3.34,
"learning_rate": 0.001,
"loss": 0.0721,
"step": 456
},
{
"epoch": 3.36,
"learning_rate": 0.001,
"loss": 0.0737,
"step": 458
},
{
"epoch": 3.37,
"learning_rate": 0.001,
"loss": 0.0717,
"step": 460
},
{
"epoch": 3.39,
"learning_rate": 0.001,
"loss": 0.0729,
"step": 462
},
{
"epoch": 3.4,
"learning_rate": 0.001,
"loss": 0.0722,
"step": 464
},
{
"epoch": 3.42,
"learning_rate": 0.001,
"loss": 0.0745,
"step": 466
},
{
"epoch": 3.43,
"learning_rate": 0.001,
"loss": 0.0779,
"step": 468
},
{
"epoch": 3.45,
"learning_rate": 0.001,
"loss": 0.0793,
"step": 470
},
{
"epoch": 3.46,
"learning_rate": 0.001,
"loss": 0.0836,
"step": 472
},
{
"epoch": 3.48,
"learning_rate": 0.001,
"loss": 0.0796,
"step": 474
},
{
"epoch": 3.49,
"learning_rate": 0.001,
"loss": 0.0806,
"step": 476
},
{
"epoch": 3.5,
"learning_rate": 0.001,
"loss": 0.0768,
"step": 478
},
{
"epoch": 3.52,
"learning_rate": 0.001,
"loss": 0.0765,
"step": 480
},
{
"epoch": 3.53,
"learning_rate": 0.001,
"loss": 0.0757,
"step": 482
},
{
"epoch": 3.55,
"learning_rate": 0.001,
"loss": 0.075,
"step": 484
},
{
"epoch": 3.56,
"learning_rate": 0.001,
"loss": 0.0735,
"step": 486
},
{
"epoch": 3.58,
"learning_rate": 0.001,
"loss": 0.0802,
"step": 488
},
{
"epoch": 3.59,
"learning_rate": 0.001,
"loss": 0.0768,
"step": 490
},
{
"epoch": 3.61,
"learning_rate": 0.001,
"loss": 0.0789,
"step": 492
},
{
"epoch": 3.62,
"learning_rate": 0.001,
"loss": 0.0801,
"step": 494
},
{
"epoch": 3.64,
"learning_rate": 0.001,
"loss": 0.0818,
"step": 496
},
{
"epoch": 3.65,
"learning_rate": 0.001,
"loss": 0.0838,
"step": 498
},
{
"epoch": 3.67,
"learning_rate": 0.001,
"loss": 0.0821,
"step": 500
},
{
"epoch": 3.68,
"learning_rate": 0.001,
"loss": 0.0831,
"step": 502
},
{
"epoch": 3.7,
"learning_rate": 0.001,
"loss": 0.0805,
"step": 504
},
{
"epoch": 3.71,
"learning_rate": 0.001,
"loss": 0.0793,
"step": 506
},
{
"epoch": 3.72,
"learning_rate": 0.001,
"loss": 0.0826,
"step": 508
},
{
"epoch": 3.74,
"learning_rate": 0.001,
"loss": 0.0825,
"step": 510
},
{
"epoch": 3.75,
"learning_rate": 0.001,
"loss": 0.0847,
"step": 512
},
{
"epoch": 3.77,
"learning_rate": 0.001,
"loss": 0.0862,
"step": 514
},
{
"epoch": 3.78,
"learning_rate": 0.001,
"loss": 0.0827,
"step": 516
},
{
"epoch": 3.8,
"learning_rate": 0.001,
"loss": 0.084,
"step": 518
},
{
"epoch": 3.81,
"learning_rate": 0.001,
"loss": 0.0854,
"step": 520
},
{
"epoch": 3.83,
"learning_rate": 0.001,
"loss": 0.0843,
"step": 522
},
{
"epoch": 3.84,
"learning_rate": 0.001,
"loss": 0.0817,
"step": 524
},
{
"epoch": 3.86,
"learning_rate": 0.001,
"loss": 0.081,
"step": 526
},
{
"epoch": 3.87,
"learning_rate": 0.001,
"loss": 0.0842,
"step": 528
},
{
"epoch": 3.89,
"learning_rate": 0.001,
"loss": 0.0798,
"step": 530
},
{
"epoch": 3.9,
"learning_rate": 0.001,
"loss": 0.0864,
"step": 532
},
{
"epoch": 3.92,
"learning_rate": 0.001,
"loss": 0.0856,
"step": 534
},
{
"epoch": 3.93,
"learning_rate": 0.001,
"loss": 0.0831,
"step": 536
},
{
"epoch": 3.94,
"learning_rate": 0.001,
"loss": 0.0844,
"step": 538
},
{
"epoch": 3.96,
"learning_rate": 0.001,
"loss": 0.0848,
"step": 540
},
{
"epoch": 3.97,
"learning_rate": 0.001,
"loss": 0.0862,
"step": 542
},
{
"epoch": 3.99,
"learning_rate": 0.001,
"loss": 0.0818,
"step": 544
},
{
"epoch": 4.0,
"eval_gen_len": 1741.0236382322714,
"eval_loss": 4.169858932495117,
"eval_rouge1": 40.2374,
"eval_rouge2": 13.5869,
"eval_rougeL": 17.364,
"eval_rougeLsum": 38.2969,
"eval_runtime": 19669.3949,
"eval_samples_per_second": 0.049,
"eval_steps_per_second": 0.006,
"step": 545
},
{
"epoch": 4.0,
"learning_rate": 0.001,
"loss": 0.0813,
"step": 546
},
{
"epoch": 4.02,
"learning_rate": 0.001,
"loss": 0.06,
"step": 548
},
{
"epoch": 4.03,
"learning_rate": 0.001,
"loss": 0.0593,
"step": 550
},
{
"epoch": 4.05,
"learning_rate": 0.001,
"loss": 0.0568,
"step": 552
},
{
"epoch": 4.06,
"learning_rate": 0.001,
"loss": 0.0591,
"step": 554
},
{
"epoch": 4.08,
"learning_rate": 0.001,
"loss": 0.0603,
"step": 556
},
{
"epoch": 4.09,
"learning_rate": 0.001,
"loss": 0.062,
"step": 558
},
{
"epoch": 4.11,
"learning_rate": 0.001,
"loss": 0.0613,
"step": 560
},
{
"epoch": 4.12,
"learning_rate": 0.001,
"loss": 0.0623,
"step": 562
},
{
"epoch": 4.14,
"learning_rate": 0.001,
"loss": 0.0625,
"step": 564
},
{
"epoch": 4.15,
"learning_rate": 0.001,
"loss": 0.0631,
"step": 566
},
{
"epoch": 4.16,
"learning_rate": 0.001,
"loss": 0.0605,
"step": 568
},
{
"epoch": 4.18,
"learning_rate": 0.001,
"loss": 0.0609,
"step": 570
},
{
"epoch": 4.19,
"learning_rate": 0.001,
"loss": 0.0623,
"step": 572
},
{
"epoch": 4.21,
"learning_rate": 0.001,
"loss": 0.0631,
"step": 574
},
{
"epoch": 4.22,
"learning_rate": 0.001,
"loss": 0.0618,
"step": 576
},
{
"epoch": 4.24,
"learning_rate": 0.001,
"loss": 0.0634,
"step": 578
},
{
"epoch": 4.25,
"learning_rate": 0.001,
"loss": 0.0623,
"step": 580
},
{
"epoch": 4.27,
"learning_rate": 0.001,
"loss": 0.0618,
"step": 582
},
{
"epoch": 4.28,
"learning_rate": 0.001,
"loss": 0.0651,
"step": 584
},
{
"epoch": 4.3,
"learning_rate": 0.001,
"loss": 0.064,
"step": 586
},
{
"epoch": 4.31,
"learning_rate": 0.001,
"loss": 0.0666,
"step": 588
},
{
"epoch": 4.33,
"learning_rate": 0.001,
"loss": 0.0644,
"step": 590
},
{
"epoch": 4.34,
"learning_rate": 0.001,
"loss": 0.0678,
"step": 592
},
{
"epoch": 4.36,
"learning_rate": 0.001,
"loss": 0.0675,
"step": 594
},
{
"epoch": 4.37,
"learning_rate": 0.001,
"loss": 0.0635,
"step": 596
},
{
"epoch": 4.38,
"learning_rate": 0.001,
"loss": 0.0669,
"step": 598
},
{
"epoch": 4.4,
"learning_rate": 0.001,
"loss": 0.0644,
"step": 600
},
{
"epoch": 4.41,
"learning_rate": 0.001,
"loss": 0.0671,
"step": 602
},
{
"epoch": 4.43,
"learning_rate": 0.001,
"loss": 0.0646,
"step": 604
},
{
"epoch": 4.44,
"learning_rate": 0.001,
"loss": 0.067,
"step": 606
},
{
"epoch": 4.46,
"learning_rate": 0.001,
"loss": 0.0663,
"step": 608
},
{
"epoch": 4.47,
"learning_rate": 0.001,
"loss": 0.0664,
"step": 610
},
{
"epoch": 4.49,
"learning_rate": 0.001,
"loss": 0.0697,
"step": 612
},
{
"epoch": 4.5,
"learning_rate": 0.001,
"loss": 0.0671,
"step": 614
},
{
"epoch": 4.52,
"learning_rate": 0.001,
"loss": 0.0701,
"step": 616
},
{
"epoch": 4.53,
"learning_rate": 0.001,
"loss": 0.0745,
"step": 618
},
{
"epoch": 4.55,
"learning_rate": 0.001,
"loss": 0.076,
"step": 620
},
{
"epoch": 4.56,
"learning_rate": 0.001,
"loss": 0.0692,
"step": 622
},
{
"epoch": 4.58,
"learning_rate": 0.001,
"loss": 0.071,
"step": 624
},
{
"epoch": 4.59,
"learning_rate": 0.001,
"loss": 0.0718,
"step": 626
},
{
"epoch": 4.6,
"learning_rate": 0.001,
"loss": 0.0731,
"step": 628
},
{
"epoch": 4.62,
"learning_rate": 0.001,
"loss": 0.078,
"step": 630
},
{
"epoch": 4.63,
"learning_rate": 0.001,
"loss": 0.0747,
"step": 632
},
{
"epoch": 4.65,
"learning_rate": 0.001,
"loss": 0.0787,
"step": 634
},
{
"epoch": 4.66,
"learning_rate": 0.001,
"loss": 0.073,
"step": 636
},
{
"epoch": 4.68,
"learning_rate": 0.001,
"loss": 0.0689,
"step": 638
},
{
"epoch": 4.69,
"learning_rate": 0.001,
"loss": 0.0722,
"step": 640
},
{
"epoch": 4.71,
"learning_rate": 0.001,
"loss": 0.0754,
"step": 642
},
{
"epoch": 4.72,
"learning_rate": 0.001,
"loss": 0.073,
"step": 644
},
{
"epoch": 4.74,
"learning_rate": 0.001,
"loss": 0.076,
"step": 646
},
{
"epoch": 4.75,
"learning_rate": 0.001,
"loss": 0.0704,
"step": 648
},
{
"epoch": 4.77,
"learning_rate": 0.001,
"loss": 0.0707,
"step": 650
},
{
"epoch": 4.78,
"learning_rate": 0.001,
"loss": 0.0707,
"step": 652
},
{
"epoch": 4.8,
"learning_rate": 0.001,
"loss": 0.071,
"step": 654
},
{
"epoch": 4.81,
"learning_rate": 0.001,
"loss": 0.0715,
"step": 656
},
{
"epoch": 4.82,
"learning_rate": 0.001,
"loss": 0.0714,
"step": 658
},
{
"epoch": 4.84,
"learning_rate": 0.001,
"loss": 0.0706,
"step": 660
},
{
"epoch": 4.85,
"learning_rate": 0.001,
"loss": 0.0718,
"step": 662
},
{
"epoch": 4.87,
"learning_rate": 0.001,
"loss": 0.0712,
"step": 664
},
{
"epoch": 4.88,
"learning_rate": 0.001,
"loss": 0.0737,
"step": 666
},
{
"epoch": 4.9,
"learning_rate": 0.001,
"loss": 0.0764,
"step": 668
},
{
"epoch": 4.91,
"learning_rate": 0.001,
"loss": 0.0766,
"step": 670
},
{
"epoch": 4.93,
"learning_rate": 0.001,
"loss": 0.0748,
"step": 672
},
{
"epoch": 4.94,
"learning_rate": 0.001,
"loss": 0.0798,
"step": 674
},
{
"epoch": 4.96,
"learning_rate": 0.001,
"loss": 0.079,
"step": 676
},
{
"epoch": 4.97,
"learning_rate": 0.001,
"loss": 0.0765,
"step": 678
},
{
"epoch": 4.99,
"learning_rate": 0.001,
"loss": 0.0786,
"step": 680
},
{
"epoch": 4.99,
"eval_gen_len": 1710.4110996916752,
"eval_loss": 4.333888530731201,
"eval_rouge1": 39.5612,
"eval_rouge2": 13.4283,
"eval_rougeL": 17.3666,
"eval_rougeLsum": 37.6526,
"eval_runtime": 19644.4491,
"eval_samples_per_second": 0.05,
"eval_steps_per_second": 0.006,
"step": 680
},
{
"epoch": 4.99,
"step": 680,
"total_flos": 1.2370900624117678e+18,
"train_loss": 0.10004215405925232,
"train_runtime": 141879.0284,
"train_samples_per_second": 0.615,
"train_steps_per_second": 0.005
}
],
"logging_steps": 2,
"max_steps": 680,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 1.2370900624117678e+18,
"trial_name": null,
"trial_params": null
}