pegasus-x-large-book-summary / trainer_state.json
pszemraj's picture
+2 epochs finetuning @ 16384 tokens input
6ace57a
raw
history blame
20 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.9969788519637461,
"global_step": 330,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 8e-05,
"loss": 2.1157,
"step": 2
},
{
"epoch": 0.02,
"learning_rate": 0.00016,
"loss": 2.2075,
"step": 4
},
{
"epoch": 0.04,
"learning_rate": 0.00024,
"loss": 2.1208,
"step": 6
},
{
"epoch": 0.05,
"learning_rate": 0.00032,
"loss": 2.2187,
"step": 8
},
{
"epoch": 0.06,
"learning_rate": 0.0004,
"loss": 2.2193,
"step": 10
},
{
"epoch": 0.07,
"learning_rate": 0.000399961448096413,
"loss": 2.1887,
"step": 12
},
{
"epoch": 0.08,
"learning_rate": 0.0003998458072481446,
"loss": 2.1639,
"step": 14
},
{
"epoch": 0.1,
"learning_rate": 0.00039965312203694324,
"loss": 2.1191,
"step": 16
},
{
"epoch": 0.11,
"learning_rate": 0.0003993834667466256,
"loss": 2.1907,
"step": 18
},
{
"epoch": 0.12,
"learning_rate": 0.0003990369453344394,
"loss": 2.1976,
"step": 20
},
{
"epoch": 0.13,
"learning_rate": 0.0003986136913909853,
"loss": 2.163,
"step": 22
},
{
"epoch": 0.15,
"learning_rate": 0.00039811386808871546,
"loss": 2.2323,
"step": 24
},
{
"epoch": 0.16,
"learning_rate": 0.00039753766811902755,
"loss": 2.1354,
"step": 26
},
{
"epoch": 0.17,
"learning_rate": 0.00039688531361797835,
"loss": 2.1643,
"step": 28
},
{
"epoch": 0.18,
"learning_rate": 0.0003961570560806461,
"loss": 2.2129,
"step": 30
},
{
"epoch": 0.19,
"learning_rate": 0.0003953531762641745,
"loss": 2.1702,
"step": 32
},
{
"epoch": 0.21,
"learning_rate": 0.0003944739840795353,
"loss": 2.1848,
"step": 34
},
{
"epoch": 0.22,
"learning_rate": 0.00039351981847205196,
"loss": 2.1619,
"step": 36
},
{
"epoch": 0.23,
"learning_rate": 0.00039249104729072946,
"loss": 2.1781,
"step": 38
},
{
"epoch": 0.24,
"learning_rate": 0.0003913880671464418,
"loss": 2.2253,
"step": 40
},
{
"epoch": 0.25,
"learning_rate": 0.00039021130325903074,
"loss": 2.1372,
"step": 42
},
{
"epoch": 0.27,
"learning_rate": 0.00038896120929337566,
"loss": 2.1798,
"step": 44
},
{
"epoch": 0.28,
"learning_rate": 0.00038763826718449685,
"loss": 2.1592,
"step": 46
},
{
"epoch": 0.29,
"learning_rate": 0.00038624298695176073,
"loss": 2.1695,
"step": 48
},
{
"epoch": 0.3,
"learning_rate": 0.0003847759065022574,
"loss": 2.2302,
"step": 50
},
{
"epoch": 0.31,
"learning_rate": 0.00038323759142342724,
"loss": 2.2537,
"step": 52
},
{
"epoch": 0.33,
"learning_rate": 0.0003816286347650163,
"loss": 2.2121,
"step": 54
},
{
"epoch": 0.34,
"learning_rate": 0.00037994965681044433,
"loss": 2.1516,
"step": 56
},
{
"epoch": 0.35,
"learning_rate": 0.0003782013048376736,
"loss": 2.1707,
"step": 58
},
{
"epoch": 0.36,
"learning_rate": 0.000376384252869671,
"loss": 2.202,
"step": 60
},
{
"epoch": 0.37,
"learning_rate": 0.00037449920141455944,
"loss": 2.1967,
"step": 62
},
{
"epoch": 0.39,
"learning_rate": 0.0003725468771955584,
"loss": 2.1701,
"step": 64
},
{
"epoch": 0.4,
"learning_rate": 0.00037052803287081844,
"loss": 2.1082,
"step": 66
},
{
"epoch": 0.41,
"learning_rate": 0.00036844344674325733,
"loss": 2.151,
"step": 68
},
{
"epoch": 0.42,
"learning_rate": 0.0003662939224605091,
"loss": 2.2102,
"step": 70
},
{
"epoch": 0.44,
"learning_rate": 0.0003640802887051027,
"loss": 2.187,
"step": 72
},
{
"epoch": 0.45,
"learning_rate": 0.0003618033988749895,
"loss": 2.2118,
"step": 74
},
{
"epoch": 0.46,
"learning_rate": 0.00035946413075454146,
"loss": 2.2232,
"step": 76
},
{
"epoch": 0.47,
"learning_rate": 0.00035706338617614897,
"loss": 2.1897,
"step": 78
},
{
"epoch": 0.48,
"learning_rate": 0.0003546020906725474,
"loss": 2.2366,
"step": 80
},
{
"epoch": 0.5,
"learning_rate": 0.0003520811931200062,
"loss": 2.1615,
"step": 82
},
{
"epoch": 0.51,
"learning_rate": 0.0003495016653725194,
"loss": 2.1765,
"step": 84
},
{
"epoch": 0.52,
"learning_rate": 0.0003468645018871371,
"loss": 2.2205,
"step": 86
},
{
"epoch": 0.53,
"learning_rate": 0.00034417071934058377,
"loss": 2.1722,
"step": 88
},
{
"epoch": 0.54,
"learning_rate": 0.0003414213562373095,
"loss": 2.1726,
"step": 90
},
{
"epoch": 0.56,
"learning_rate": 0.0003386174725091272,
"loss": 2.1401,
"step": 92
},
{
"epoch": 0.57,
"learning_rate": 0.0003357601491065884,
"loss": 2.2254,
"step": 94
},
{
"epoch": 0.58,
"learning_rate": 0.00033285048758225635,
"loss": 2.1489,
"step": 96
},
{
"epoch": 0.59,
"learning_rate": 0.0003298896096660367,
"loss": 2.2459,
"step": 98
},
{
"epoch": 0.6,
"learning_rate": 0.0003268786568327291,
"loss": 2.2103,
"step": 100
},
{
"epoch": 0.62,
"learning_rate": 0.00032381878986196687,
"loss": 2.1864,
"step": 102
},
{
"epoch": 0.63,
"learning_rate": 0.0003207111883907143,
"loss": 2.1641,
"step": 104
},
{
"epoch": 0.64,
"learning_rate": 0.00031755705045849464,
"loss": 2.1346,
"step": 106
},
{
"epoch": 0.65,
"learning_rate": 0.00031435759204552244,
"loss": 2.1639,
"step": 108
},
{
"epoch": 0.66,
"learning_rate": 0.00031111404660392046,
"loss": 2.2032,
"step": 110
},
{
"epoch": 0.68,
"learning_rate": 0.00030782766458220014,
"loss": 2.2095,
"step": 112
},
{
"epoch": 0.69,
"learning_rate": 0.0003044997129431898,
"loss": 2.1789,
"step": 114
},
{
"epoch": 0.7,
"learning_rate": 0.00030113147467559695,
"loss": 2.1947,
"step": 116
},
{
"epoch": 0.71,
"learning_rate": 0.00029772424829939106,
"loss": 2.2011,
"step": 118
},
{
"epoch": 0.73,
"learning_rate": 0.0002942793473651996,
"loss": 2.1956,
"step": 120
},
{
"epoch": 0.74,
"learning_rate": 0.00029079809994790937,
"loss": 2.2043,
"step": 122
},
{
"epoch": 0.75,
"learning_rate": 0.0002872818481346684,
"loss": 2.1571,
"step": 124
},
{
"epoch": 0.76,
"learning_rate": 0.0002837319475074856,
"loss": 2.2016,
"step": 126
},
{
"epoch": 0.77,
"learning_rate": 0.0002801497666206282,
"loss": 2.182,
"step": 128
},
{
"epoch": 0.79,
"learning_rate": 0.000276536686473018,
"loss": 2.1583,
"step": 130
},
{
"epoch": 0.8,
"learning_rate": 0.00027289409997583,
"loss": 2.2161,
"step": 132
},
{
"epoch": 0.81,
"learning_rate": 0.0002692234114154986,
"loss": 2.2267,
"step": 134
},
{
"epoch": 0.82,
"learning_rate": 0.00026552603591233873,
"loss": 2.2418,
"step": 136
},
{
"epoch": 0.83,
"learning_rate": 0.00026180339887498953,
"loss": 2.1828,
"step": 138
},
{
"epoch": 0.85,
"learning_rate": 0.00025805693545089247,
"loss": 2.1103,
"step": 140
},
{
"epoch": 0.86,
"learning_rate": 0.00025428808997301485,
"loss": 2.2022,
"step": 142
},
{
"epoch": 0.87,
"learning_rate": 0.0002504983154030316,
"loss": 2.1691,
"step": 144
},
{
"epoch": 0.88,
"learning_rate": 0.00024668907277118114,
"loss": 2.1787,
"step": 146
},
{
"epoch": 0.89,
"learning_rate": 0.00024286183061301015,
"loss": 2.2069,
"step": 148
},
{
"epoch": 0.91,
"learning_rate": 0.0002390180644032257,
"loss": 2.2609,
"step": 150
},
{
"epoch": 0.92,
"learning_rate": 0.00023515925598687094,
"loss": 2.2089,
"step": 152
},
{
"epoch": 0.93,
"learning_rate": 0.0002312868930080462,
"loss": 2.1889,
"step": 154
},
{
"epoch": 0.94,
"learning_rate": 0.00022740246833639365,
"loss": 2.2167,
"step": 156
},
{
"epoch": 0.95,
"learning_rate": 0.00022350747949156756,
"loss": 2.1636,
"step": 158
},
{
"epoch": 0.97,
"learning_rate": 0.0002196034280659122,
"loss": 2.2215,
"step": 160
},
{
"epoch": 0.98,
"learning_rate": 0.000215691819145569,
"loss": 2.1903,
"step": 162
},
{
"epoch": 0.99,
"learning_rate": 0.0002117741607302378,
"loss": 2.2131,
"step": 164
},
{
"epoch": 1.01,
"learning_rate": 0.00020785196315181376,
"loss": 2.6465,
"step": 166
},
{
"epoch": 1.02,
"learning_rate": 0.00020392673849212566,
"loss": 2.18,
"step": 168
},
{
"epoch": 1.03,
"learning_rate": 0.0002,
"loss": 2.0911,
"step": 170
},
{
"epoch": 1.04,
"learning_rate": 0.00019607326150787436,
"loss": 2.1044,
"step": 172
},
{
"epoch": 1.05,
"learning_rate": 0.00019214803684818634,
"loss": 2.1266,
"step": 174
},
{
"epoch": 1.07,
"learning_rate": 0.00018822583926976218,
"loss": 2.1559,
"step": 176
},
{
"epoch": 1.08,
"learning_rate": 0.00018430818085443104,
"loss": 2.1522,
"step": 178
},
{
"epoch": 1.09,
"learning_rate": 0.00018039657193408788,
"loss": 2.1535,
"step": 180
},
{
"epoch": 1.1,
"learning_rate": 0.00017649252050843252,
"loss": 2.2344,
"step": 182
},
{
"epoch": 1.11,
"learning_rate": 0.00017259753166360642,
"loss": 2.121,
"step": 184
},
{
"epoch": 1.13,
"learning_rate": 0.00016871310699195379,
"loss": 2.1426,
"step": 186
},
{
"epoch": 1.14,
"learning_rate": 0.0001648407440131291,
"loss": 2.1322,
"step": 188
},
{
"epoch": 1.15,
"learning_rate": 0.00016098193559677438,
"loss": 2.1529,
"step": 190
},
{
"epoch": 1.16,
"learning_rate": 0.0001571381693869899,
"loss": 2.18,
"step": 192
},
{
"epoch": 1.18,
"learning_rate": 0.000153310927228819,
"loss": 2.1437,
"step": 194
},
{
"epoch": 1.19,
"learning_rate": 0.0001495016845969684,
"loss": 2.1725,
"step": 196
},
{
"epoch": 1.2,
"learning_rate": 0.00014571191002698517,
"loss": 2.1208,
"step": 198
},
{
"epoch": 1.21,
"learning_rate": 0.00014194306454910757,
"loss": 2.1427,
"step": 200
},
{
"epoch": 1.22,
"learning_rate": 0.00013819660112501054,
"loss": 2.1056,
"step": 202
},
{
"epoch": 1.24,
"learning_rate": 0.00013447396408766132,
"loss": 2.1255,
"step": 204
},
{
"epoch": 1.25,
"learning_rate": 0.00013077658858450138,
"loss": 2.1658,
"step": 206
},
{
"epoch": 1.26,
"learning_rate": 0.00012710590002417007,
"loss": 2.1443,
"step": 208
},
{
"epoch": 1.27,
"learning_rate": 0.00012346331352698205,
"loss": 2.1308,
"step": 210
},
{
"epoch": 1.28,
"learning_rate": 0.00011985023337937184,
"loss": 2.1563,
"step": 212
},
{
"epoch": 1.3,
"learning_rate": 0.00011626805249251445,
"loss": 2.1129,
"step": 214
},
{
"epoch": 1.31,
"learning_rate": 0.00011271815186533155,
"loss": 2.1279,
"step": 216
},
{
"epoch": 1.32,
"learning_rate": 0.00010920190005209065,
"loss": 2.0998,
"step": 218
},
{
"epoch": 1.33,
"learning_rate": 0.00010572065263480046,
"loss": 2.1541,
"step": 220
},
{
"epoch": 1.34,
"learning_rate": 0.00010227575170060909,
"loss": 2.162,
"step": 222
},
{
"epoch": 1.36,
"learning_rate": 9.886852532440312e-05,
"loss": 2.1697,
"step": 224
},
{
"epoch": 1.37,
"learning_rate": 9.550028705681025e-05,
"loss": 2.1428,
"step": 226
},
{
"epoch": 1.38,
"learning_rate": 9.217233541779995e-05,
"loss": 2.0846,
"step": 228
},
{
"epoch": 1.39,
"learning_rate": 8.888595339607961e-05,
"loss": 2.136,
"step": 230
},
{
"epoch": 1.4,
"learning_rate": 8.564240795447758e-05,
"loss": 2.2081,
"step": 232
},
{
"epoch": 1.42,
"learning_rate": 8.24429495415054e-05,
"loss": 2.097,
"step": 234
},
{
"epoch": 1.43,
"learning_rate": 7.928881160928572e-05,
"loss": 2.1502,
"step": 236
},
{
"epoch": 1.44,
"learning_rate": 7.618121013803319e-05,
"loss": 2.1683,
"step": 238
},
{
"epoch": 1.45,
"learning_rate": 7.312134316727093e-05,
"loss": 2.1311,
"step": 240
},
{
"epoch": 1.47,
"learning_rate": 7.011039033396329e-05,
"loss": 2.15,
"step": 242
},
{
"epoch": 1.48,
"learning_rate": 6.714951241774372e-05,
"loss": 2.1618,
"step": 244
},
{
"epoch": 1.49,
"learning_rate": 6.423985089341164e-05,
"loss": 2.1256,
"step": 246
},
{
"epoch": 1.5,
"learning_rate": 6.138252749087286e-05,
"loss": 2.194,
"step": 248
},
{
"epoch": 1.51,
"learning_rate": 5.857864376269051e-05,
"loss": 2.1898,
"step": 250
},
{
"epoch": 1.53,
"learning_rate": 5.5829280659416237e-05,
"loss": 2.1065,
"step": 252
},
{
"epoch": 1.54,
"learning_rate": 5.313549811286293e-05,
"loss": 2.0985,
"step": 254
},
{
"epoch": 1.55,
"learning_rate": 5.0498334627480615e-05,
"loss": 2.1658,
"step": 256
},
{
"epoch": 1.56,
"learning_rate": 4.7918806879993814e-05,
"loss": 2.1722,
"step": 258
},
{
"epoch": 1.57,
"learning_rate": 4.53979093274526e-05,
"loss": 2.083,
"step": 260
},
{
"epoch": 1.59,
"learning_rate": 4.293661382385106e-05,
"loss": 2.1665,
"step": 262
},
{
"epoch": 1.6,
"learning_rate": 4.05358692454586e-05,
"loss": 2.133,
"step": 264
},
{
"epoch": 1.61,
"learning_rate": 3.819660112501053e-05,
"loss": 2.1471,
"step": 266
},
{
"epoch": 1.62,
"learning_rate": 3.591971129489728e-05,
"loss": 2.1791,
"step": 268
},
{
"epoch": 1.63,
"learning_rate": 3.370607753949093e-05,
"loss": 2.072,
"step": 270
},
{
"epoch": 1.65,
"learning_rate": 3.1556553256742713e-05,
"loss": 2.1349,
"step": 272
},
{
"epoch": 1.66,
"learning_rate": 2.9471967129181565e-05,
"loss": 2.1705,
"step": 274
},
{
"epoch": 1.67,
"learning_rate": 2.7453122804441634e-05,
"loss": 2.1388,
"step": 276
},
{
"epoch": 1.68,
"learning_rate": 2.5500798585440567e-05,
"loss": 2.1586,
"step": 278
},
{
"epoch": 1.69,
"learning_rate": 2.3615747130329013e-05,
"loss": 2.1193,
"step": 280
},
{
"epoch": 1.71,
"learning_rate": 2.1798695162326442e-05,
"loss": 2.0693,
"step": 282
},
{
"epoch": 1.72,
"learning_rate": 2.0050343189555743e-05,
"loss": 2.1118,
"step": 284
},
{
"epoch": 1.73,
"learning_rate": 1.8371365234983727e-05,
"loss": 2.1395,
"step": 286
},
{
"epoch": 1.74,
"learning_rate": 1.676240857657283e-05,
"loss": 2.1948,
"step": 288
},
{
"epoch": 1.76,
"learning_rate": 1.5224093497742653e-05,
"loss": 2.1251,
"step": 290
},
{
"epoch": 1.77,
"learning_rate": 1.3757013048239286e-05,
"loss": 2.0808,
"step": 292
},
{
"epoch": 1.78,
"learning_rate": 1.236173281550319e-05,
"loss": 2.1342,
"step": 294
},
{
"epoch": 1.79,
"learning_rate": 1.1038790706624391e-05,
"loss": 2.0986,
"step": 296
},
{
"epoch": 1.8,
"learning_rate": 9.788696740969295e-06,
"loss": 2.123,
"step": 298
},
{
"epoch": 1.82,
"learning_rate": 8.611932853558236e-06,
"loss": 2.094,
"step": 300
},
{
"epoch": 1.83,
"learning_rate": 7.508952709270567e-06,
"loss": 2.1003,
"step": 302
},
{
"epoch": 1.84,
"learning_rate": 6.480181527948048e-06,
"loss": 2.188,
"step": 304
},
{
"epoch": 1.85,
"learning_rate": 5.5260159204646885e-06,
"loss": 2.1912,
"step": 306
},
{
"epoch": 1.86,
"learning_rate": 4.646823735825523e-06,
"loss": 2.1468,
"step": 308
},
{
"epoch": 1.88,
"learning_rate": 3.842943919353914e-06,
"loss": 2.1543,
"step": 310
},
{
"epoch": 1.89,
"learning_rate": 3.114686382021681e-06,
"loss": 2.0745,
"step": 312
},
{
"epoch": 1.9,
"learning_rate": 2.462331880972468e-06,
"loss": 2.1252,
"step": 314
},
{
"epoch": 1.91,
"learning_rate": 1.88613191128455e-06,
"loss": 2.1826,
"step": 316
},
{
"epoch": 1.92,
"learning_rate": 1.3863086090147415e-06,
"loss": 2.159,
"step": 318
},
{
"epoch": 1.94,
"learning_rate": 9.630546655606364e-07,
"loss": 2.1398,
"step": 320
},
{
"epoch": 1.95,
"learning_rate": 6.165332533744073e-07,
"loss": 2.1637,
"step": 322
},
{
"epoch": 1.96,
"learning_rate": 3.468779630568353e-07,
"loss": 2.0743,
"step": 324
},
{
"epoch": 1.97,
"learning_rate": 1.5419275185541982e-07,
"loss": 2.1162,
"step": 326
},
{
"epoch": 1.98,
"learning_rate": 3.855190358703631e-08,
"loss": 2.1847,
"step": 328
},
{
"epoch": 2.0,
"learning_rate": 0.0,
"loss": 2.092,
"step": 330
}
],
"max_steps": 330,
"num_train_epochs": 2,
"total_flos": 9.774541129058877e+17,
"trial_name": null,
"trial_params": null
}