hushell's picture
Model save
b2f9a13 verified
{
"best_metric": 2.10353946685791,
"best_model_checkpoint": "data/tinyllama_mole_sft_router05_lr1e-4_ep3/checkpoint-2200",
"epoch": 2.9986859395532193,
"eval_steps": 100,
"global_step": 3423,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 8.333333333333333e-07,
"loss": 3.7925,
"step": 1
},
{
"epoch": 0.0,
"learning_rate": 4.166666666666667e-06,
"loss": 3.7302,
"step": 5
},
{
"epoch": 0.01,
"learning_rate": 8.333333333333334e-06,
"loss": 3.3746,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 1.25e-05,
"loss": 2.8214,
"step": 15
},
{
"epoch": 0.02,
"learning_rate": 1.6666666666666667e-05,
"loss": 2.6367,
"step": 20
},
{
"epoch": 0.02,
"learning_rate": 2.0833333333333336e-05,
"loss": 2.5137,
"step": 25
},
{
"epoch": 0.03,
"learning_rate": 2.5e-05,
"loss": 2.4115,
"step": 30
},
{
"epoch": 0.03,
"learning_rate": 2.916666666666667e-05,
"loss": 2.3625,
"step": 35
},
{
"epoch": 0.04,
"learning_rate": 3.3333333333333335e-05,
"loss": 2.3396,
"step": 40
},
{
"epoch": 0.04,
"learning_rate": 3.7500000000000003e-05,
"loss": 2.3126,
"step": 45
},
{
"epoch": 0.04,
"learning_rate": 4.166666666666667e-05,
"loss": 2.3046,
"step": 50
},
{
"epoch": 0.05,
"learning_rate": 4.5833333333333334e-05,
"loss": 2.2724,
"step": 55
},
{
"epoch": 0.05,
"learning_rate": 5e-05,
"loss": 2.2881,
"step": 60
},
{
"epoch": 0.06,
"learning_rate": 5.4166666666666664e-05,
"loss": 2.2671,
"step": 65
},
{
"epoch": 0.06,
"learning_rate": 5.833333333333334e-05,
"loss": 2.2738,
"step": 70
},
{
"epoch": 0.07,
"learning_rate": 6.25e-05,
"loss": 2.2463,
"step": 75
},
{
"epoch": 0.07,
"learning_rate": 6.666666666666667e-05,
"loss": 2.2509,
"step": 80
},
{
"epoch": 0.07,
"learning_rate": 7.083333333333334e-05,
"loss": 2.2554,
"step": 85
},
{
"epoch": 0.08,
"learning_rate": 7.500000000000001e-05,
"loss": 2.2522,
"step": 90
},
{
"epoch": 0.08,
"learning_rate": 7.916666666666666e-05,
"loss": 2.2278,
"step": 95
},
{
"epoch": 0.09,
"learning_rate": 8.333333333333334e-05,
"loss": 2.2617,
"step": 100
},
{
"epoch": 0.09,
"eval_loss": 2.2410387992858887,
"eval_runtime": 439.5124,
"eval_samples_per_second": 36.782,
"eval_steps_per_second": 1.151,
"step": 100
},
{
"epoch": 0.09,
"learning_rate": 8.75e-05,
"loss": 2.2391,
"step": 105
},
{
"epoch": 0.1,
"learning_rate": 9.166666666666667e-05,
"loss": 2.2334,
"step": 110
},
{
"epoch": 0.1,
"learning_rate": 9.583333333333334e-05,
"loss": 2.2407,
"step": 115
},
{
"epoch": 0.11,
"learning_rate": 0.0001,
"loss": 2.2282,
"step": 120
},
{
"epoch": 0.11,
"learning_rate": 9.999943459219818e-05,
"loss": 2.232,
"step": 125
},
{
"epoch": 0.11,
"learning_rate": 9.999773838158017e-05,
"loss": 2.2113,
"step": 130
},
{
"epoch": 0.12,
"learning_rate": 9.9994911406508e-05,
"loss": 2.2273,
"step": 135
},
{
"epoch": 0.12,
"learning_rate": 9.99909537309174e-05,
"loss": 2.2408,
"step": 140
},
{
"epoch": 0.13,
"learning_rate": 9.998586544431642e-05,
"loss": 2.2243,
"step": 145
},
{
"epoch": 0.13,
"learning_rate": 9.997964666178332e-05,
"loss": 2.2275,
"step": 150
},
{
"epoch": 0.14,
"learning_rate": 9.997229752396404e-05,
"loss": 2.2214,
"step": 155
},
{
"epoch": 0.14,
"learning_rate": 9.996381819706897e-05,
"loss": 2.2175,
"step": 160
},
{
"epoch": 0.14,
"learning_rate": 9.995420887286921e-05,
"loss": 2.2134,
"step": 165
},
{
"epoch": 0.15,
"learning_rate": 9.994346976869222e-05,
"loss": 2.219,
"step": 170
},
{
"epoch": 0.15,
"learning_rate": 9.993160112741697e-05,
"loss": 2.2295,
"step": 175
},
{
"epoch": 0.16,
"learning_rate": 9.991860321746833e-05,
"loss": 2.2077,
"step": 180
},
{
"epoch": 0.16,
"learning_rate": 9.990447633281108e-05,
"loss": 2.2074,
"step": 185
},
{
"epoch": 0.17,
"learning_rate": 9.988922079294327e-05,
"loss": 2.2145,
"step": 190
},
{
"epoch": 0.17,
"learning_rate": 9.987283694288896e-05,
"loss": 2.2054,
"step": 195
},
{
"epoch": 0.18,
"learning_rate": 9.985532515319038e-05,
"loss": 2.2246,
"step": 200
},
{
"epoch": 0.18,
"eval_loss": 2.21648907661438,
"eval_runtime": 475.9972,
"eval_samples_per_second": 33.962,
"eval_steps_per_second": 1.063,
"step": 200
},
{
"epoch": 0.18,
"learning_rate": 9.983668581989965e-05,
"loss": 2.2172,
"step": 205
},
{
"epoch": 0.18,
"learning_rate": 9.981691936456975e-05,
"loss": 2.1936,
"step": 210
},
{
"epoch": 0.19,
"learning_rate": 9.9796026234245e-05,
"loss": 2.2234,
"step": 215
},
{
"epoch": 0.19,
"learning_rate": 9.977400690145097e-05,
"loss": 2.2015,
"step": 220
},
{
"epoch": 0.2,
"learning_rate": 9.975086186418373e-05,
"loss": 2.1856,
"step": 225
},
{
"epoch": 0.2,
"learning_rate": 9.972659164589871e-05,
"loss": 2.2081,
"step": 230
},
{
"epoch": 0.21,
"learning_rate": 9.970119679549868e-05,
"loss": 2.2161,
"step": 235
},
{
"epoch": 0.21,
"learning_rate": 9.967467788732156e-05,
"loss": 2.1883,
"step": 240
},
{
"epoch": 0.21,
"learning_rate": 9.964703552112722e-05,
"loss": 2.2166,
"step": 245
},
{
"epoch": 0.22,
"learning_rate": 9.961827032208406e-05,
"loss": 2.1922,
"step": 250
},
{
"epoch": 0.22,
"learning_rate": 9.958838294075479e-05,
"loss": 2.2028,
"step": 255
},
{
"epoch": 0.23,
"learning_rate": 9.955737405308174e-05,
"loss": 2.2048,
"step": 260
},
{
"epoch": 0.23,
"learning_rate": 9.952524436037161e-05,
"loss": 2.1892,
"step": 265
},
{
"epoch": 0.24,
"learning_rate": 9.949199458927955e-05,
"loss": 2.1891,
"step": 270
},
{
"epoch": 0.24,
"learning_rate": 9.945762549179277e-05,
"loss": 2.2036,
"step": 275
},
{
"epoch": 0.25,
"learning_rate": 9.942213784521346e-05,
"loss": 2.1801,
"step": 280
},
{
"epoch": 0.25,
"learning_rate": 9.938553245214138e-05,
"loss": 2.2159,
"step": 285
},
{
"epoch": 0.25,
"learning_rate": 9.934781014045545e-05,
"loss": 2.1817,
"step": 290
},
{
"epoch": 0.26,
"learning_rate": 9.93089717632953e-05,
"loss": 2.2027,
"step": 295
},
{
"epoch": 0.26,
"learning_rate": 9.926901819904178e-05,
"loss": 2.1994,
"step": 300
},
{
"epoch": 0.26,
"eval_loss": 2.1994447708129883,
"eval_runtime": 435.4362,
"eval_samples_per_second": 37.126,
"eval_steps_per_second": 1.162,
"step": 300
},
{
"epoch": 0.27,
"learning_rate": 9.922795035129713e-05,
"loss": 2.199,
"step": 305
},
{
"epoch": 0.27,
"learning_rate": 9.918576914886465e-05,
"loss": 2.1699,
"step": 310
},
{
"epoch": 0.28,
"learning_rate": 9.914247554572757e-05,
"loss": 2.1873,
"step": 315
},
{
"epoch": 0.28,
"learning_rate": 9.909807052102751e-05,
"loss": 2.1832,
"step": 320
},
{
"epoch": 0.28,
"learning_rate": 9.905255507904238e-05,
"loss": 2.1833,
"step": 325
},
{
"epoch": 0.29,
"learning_rate": 9.900593024916365e-05,
"loss": 2.1766,
"step": 330
},
{
"epoch": 0.29,
"learning_rate": 9.895819708587295e-05,
"loss": 2.2268,
"step": 335
},
{
"epoch": 0.3,
"learning_rate": 9.890935666871847e-05,
"loss": 2.1813,
"step": 340
},
{
"epoch": 0.3,
"learning_rate": 9.885941010229027e-05,
"loss": 2.186,
"step": 345
},
{
"epoch": 0.31,
"learning_rate": 9.880835851619554e-05,
"loss": 2.1832,
"step": 350
},
{
"epoch": 0.31,
"learning_rate": 9.875620306503282e-05,
"loss": 2.1824,
"step": 355
},
{
"epoch": 0.32,
"learning_rate": 9.870294492836612e-05,
"loss": 2.1898,
"step": 360
},
{
"epoch": 0.32,
"learning_rate": 9.864858531069807e-05,
"loss": 2.1846,
"step": 365
},
{
"epoch": 0.32,
"learning_rate": 9.859312544144272e-05,
"loss": 2.1772,
"step": 370
},
{
"epoch": 0.33,
"learning_rate": 9.85365665748978e-05,
"loss": 2.1724,
"step": 375
},
{
"epoch": 0.33,
"learning_rate": 9.847890999021629e-05,
"loss": 2.1862,
"step": 380
},
{
"epoch": 0.34,
"learning_rate": 9.84201569913775e-05,
"loss": 2.1746,
"step": 385
},
{
"epoch": 0.34,
"learning_rate": 9.83603089071576e-05,
"loss": 2.1679,
"step": 390
},
{
"epoch": 0.35,
"learning_rate": 9.829936709109949e-05,
"loss": 2.1705,
"step": 395
},
{
"epoch": 0.35,
"learning_rate": 9.823733292148236e-05,
"loss": 2.1767,
"step": 400
},
{
"epoch": 0.35,
"eval_loss": 2.186896800994873,
"eval_runtime": 433.1626,
"eval_samples_per_second": 37.321,
"eval_steps_per_second": 1.168,
"step": 400
},
{
"epoch": 0.35,
"learning_rate": 9.817420780129031e-05,
"loss": 2.1981,
"step": 405
},
{
"epoch": 0.36,
"learning_rate": 9.810999315818077e-05,
"loss": 2.1918,
"step": 410
},
{
"epoch": 0.36,
"learning_rate": 9.804469044445215e-05,
"loss": 2.1802,
"step": 415
},
{
"epoch": 0.37,
"learning_rate": 9.797830113701101e-05,
"loss": 2.1797,
"step": 420
},
{
"epoch": 0.37,
"learning_rate": 9.791082673733864e-05,
"loss": 2.1725,
"step": 425
},
{
"epoch": 0.38,
"learning_rate": 9.784226877145712e-05,
"loss": 2.1803,
"step": 430
},
{
"epoch": 0.38,
"learning_rate": 9.77726287898948e-05,
"loss": 2.1648,
"step": 435
},
{
"epoch": 0.39,
"learning_rate": 9.770190836765123e-05,
"loss": 2.1775,
"step": 440
},
{
"epoch": 0.39,
"learning_rate": 9.763010910416155e-05,
"loss": 2.1689,
"step": 445
},
{
"epoch": 0.39,
"learning_rate": 9.755723262326031e-05,
"loss": 2.1968,
"step": 450
},
{
"epoch": 0.4,
"learning_rate": 9.748328057314476e-05,
"loss": 2.1676,
"step": 455
},
{
"epoch": 0.4,
"learning_rate": 9.740825462633752e-05,
"loss": 2.1657,
"step": 460
},
{
"epoch": 0.41,
"learning_rate": 9.733215647964884e-05,
"loss": 2.1778,
"step": 465
},
{
"epoch": 0.41,
"learning_rate": 9.725498785413813e-05,
"loss": 2.1802,
"step": 470
},
{
"epoch": 0.42,
"learning_rate": 9.717675049507514e-05,
"loss": 2.1739,
"step": 475
},
{
"epoch": 0.42,
"learning_rate": 9.709744617190038e-05,
"loss": 2.1821,
"step": 480
},
{
"epoch": 0.42,
"learning_rate": 9.701707667818515e-05,
"loss": 2.1494,
"step": 485
},
{
"epoch": 0.43,
"learning_rate": 9.693564383159103e-05,
"loss": 2.1817,
"step": 490
},
{
"epoch": 0.43,
"learning_rate": 9.685314947382868e-05,
"loss": 2.1639,
"step": 495
},
{
"epoch": 0.44,
"learning_rate": 9.676959547061624e-05,
"loss": 2.1532,
"step": 500
},
{
"epoch": 0.44,
"eval_loss": 2.179184913635254,
"eval_runtime": 432.5255,
"eval_samples_per_second": 37.376,
"eval_steps_per_second": 1.17,
"step": 500
},
{
"epoch": 0.44,
"learning_rate": 9.668498371163712e-05,
"loss": 2.1665,
"step": 505
},
{
"epoch": 0.45,
"learning_rate": 9.659931611049729e-05,
"loss": 2.1692,
"step": 510
},
{
"epoch": 0.45,
"learning_rate": 9.651259460468191e-05,
"loss": 2.1739,
"step": 515
},
{
"epoch": 0.46,
"learning_rate": 9.642482115551164e-05,
"loss": 2.1475,
"step": 520
},
{
"epoch": 0.46,
"learning_rate": 9.63359977480982e-05,
"loss": 2.1647,
"step": 525
},
{
"epoch": 0.46,
"learning_rate": 9.62461263912995e-05,
"loss": 2.159,
"step": 530
},
{
"epoch": 0.47,
"learning_rate": 9.615520911767418e-05,
"loss": 2.1532,
"step": 535
},
{
"epoch": 0.47,
"learning_rate": 9.606324798343567e-05,
"loss": 2.1686,
"step": 540
},
{
"epoch": 0.48,
"learning_rate": 9.597024506840568e-05,
"loss": 2.156,
"step": 545
},
{
"epoch": 0.48,
"learning_rate": 9.587620247596716e-05,
"loss": 2.1666,
"step": 550
},
{
"epoch": 0.49,
"learning_rate": 9.578112233301675e-05,
"loss": 2.1713,
"step": 555
},
{
"epoch": 0.49,
"learning_rate": 9.568500678991661e-05,
"loss": 2.1769,
"step": 560
},
{
"epoch": 0.49,
"learning_rate": 9.558785802044586e-05,
"loss": 2.1737,
"step": 565
},
{
"epoch": 0.5,
"learning_rate": 9.54896782217514e-05,
"loss": 2.1649,
"step": 570
},
{
"epoch": 0.5,
"learning_rate": 9.53904696142982e-05,
"loss": 2.1734,
"step": 575
},
{
"epoch": 0.51,
"learning_rate": 9.529023444181907e-05,
"loss": 2.1535,
"step": 580
},
{
"epoch": 0.51,
"learning_rate": 9.518897497126396e-05,
"loss": 2.1719,
"step": 585
},
{
"epoch": 0.52,
"learning_rate": 9.508669349274866e-05,
"loss": 2.1583,
"step": 590
},
{
"epoch": 0.52,
"learning_rate": 9.498339231950298e-05,
"loss": 2.183,
"step": 595
},
{
"epoch": 0.53,
"learning_rate": 9.487907378781853e-05,
"loss": 2.171,
"step": 600
},
{
"epoch": 0.53,
"eval_loss": 2.1716604232788086,
"eval_runtime": 434.3111,
"eval_samples_per_second": 37.222,
"eval_steps_per_second": 1.165,
"step": 600
},
{
"epoch": 0.53,
"learning_rate": 9.477374025699577e-05,
"loss": 2.143,
"step": 605
},
{
"epoch": 0.53,
"learning_rate": 9.466739410929069e-05,
"loss": 2.155,
"step": 610
},
{
"epoch": 0.54,
"learning_rate": 9.456003774986095e-05,
"loss": 2.1776,
"step": 615
},
{
"epoch": 0.54,
"learning_rate": 9.44516736067115e-05,
"loss": 2.1561,
"step": 620
},
{
"epoch": 0.55,
"learning_rate": 9.43423041306396e-05,
"loss": 2.1547,
"step": 625
},
{
"epoch": 0.55,
"learning_rate": 9.423193179517947e-05,
"loss": 2.1554,
"step": 630
},
{
"epoch": 0.56,
"learning_rate": 9.412055909654627e-05,
"loss": 2.1779,
"step": 635
},
{
"epoch": 0.56,
"learning_rate": 9.400818855357972e-05,
"loss": 2.1636,
"step": 640
},
{
"epoch": 0.57,
"learning_rate": 9.38948227076871e-05,
"loss": 2.1668,
"step": 645
},
{
"epoch": 0.57,
"learning_rate": 9.378046412278574e-05,
"loss": 2.1548,
"step": 650
},
{
"epoch": 0.57,
"learning_rate": 9.366511538524509e-05,
"loss": 2.1594,
"step": 655
},
{
"epoch": 0.58,
"learning_rate": 9.35487791038282e-05,
"loss": 2.1619,
"step": 660
},
{
"epoch": 0.58,
"learning_rate": 9.343145790963273e-05,
"loss": 2.1541,
"step": 665
},
{
"epoch": 0.59,
"learning_rate": 9.331315445603137e-05,
"loss": 2.1423,
"step": 670
},
{
"epoch": 0.59,
"learning_rate": 9.3193871418612e-05,
"loss": 2.165,
"step": 675
},
{
"epoch": 0.6,
"learning_rate": 9.3073611495117e-05,
"loss": 2.1749,
"step": 680
},
{
"epoch": 0.6,
"learning_rate": 9.295237740538232e-05,
"loss": 2.154,
"step": 685
},
{
"epoch": 0.6,
"learning_rate": 9.283017189127598e-05,
"loss": 2.1557,
"step": 690
},
{
"epoch": 0.61,
"learning_rate": 9.270699771663602e-05,
"loss": 2.1496,
"step": 695
},
{
"epoch": 0.61,
"learning_rate": 9.258285766720802e-05,
"loss": 2.1588,
"step": 700
},
{
"epoch": 0.61,
"eval_loss": 2.1645405292510986,
"eval_runtime": 433.8926,
"eval_samples_per_second": 37.258,
"eval_steps_per_second": 1.166,
"step": 700
},
{
"epoch": 0.62,
"learning_rate": 9.245775455058208e-05,
"loss": 2.1585,
"step": 705
},
{
"epoch": 0.62,
"learning_rate": 9.23316911961293e-05,
"loss": 2.1316,
"step": 710
},
{
"epoch": 0.63,
"learning_rate": 9.220467045493787e-05,
"loss": 2.1385,
"step": 715
},
{
"epoch": 0.63,
"learning_rate": 9.207669519974851e-05,
"loss": 2.1692,
"step": 720
},
{
"epoch": 0.64,
"learning_rate": 9.194776832488952e-05,
"loss": 2.1358,
"step": 725
},
{
"epoch": 0.64,
"learning_rate": 9.181789274621133e-05,
"loss": 2.1739,
"step": 730
},
{
"epoch": 0.64,
"learning_rate": 9.168707140102057e-05,
"loss": 2.1616,
"step": 735
},
{
"epoch": 0.65,
"learning_rate": 9.155530724801363e-05,
"loss": 2.1517,
"step": 740
},
{
"epoch": 0.65,
"learning_rate": 9.142260326720969e-05,
"loss": 2.1642,
"step": 745
},
{
"epoch": 0.66,
"learning_rate": 9.128896245988338e-05,
"loss": 2.1499,
"step": 750
},
{
"epoch": 0.66,
"learning_rate": 9.115438784849692e-05,
"loss": 2.1646,
"step": 755
},
{
"epoch": 0.67,
"learning_rate": 9.101888247663173e-05,
"loss": 2.1492,
"step": 760
},
{
"epoch": 0.67,
"learning_rate": 9.088244940891957e-05,
"loss": 2.1331,
"step": 765
},
{
"epoch": 0.67,
"learning_rate": 9.074509173097328e-05,
"loss": 2.137,
"step": 770
},
{
"epoch": 0.68,
"learning_rate": 9.060681254931698e-05,
"loss": 2.1401,
"step": 775
},
{
"epoch": 0.68,
"learning_rate": 9.046761499131578e-05,
"loss": 2.134,
"step": 780
},
{
"epoch": 0.69,
"learning_rate": 9.032750220510509e-05,
"loss": 2.1545,
"step": 785
},
{
"epoch": 0.69,
"learning_rate": 9.018647735951943e-05,
"loss": 2.1285,
"step": 790
},
{
"epoch": 0.7,
"learning_rate": 9.00445436440207e-05,
"loss": 2.1547,
"step": 795
},
{
"epoch": 0.7,
"learning_rate": 8.990170426862612e-05,
"loss": 2.145,
"step": 800
},
{
"epoch": 0.7,
"eval_loss": 2.156743049621582,
"eval_runtime": 439.2176,
"eval_samples_per_second": 36.806,
"eval_steps_per_second": 1.152,
"step": 800
},
{
"epoch": 0.71,
"learning_rate": 8.975796246383556e-05,
"loss": 2.128,
"step": 805
},
{
"epoch": 0.71,
"learning_rate": 8.961332148055855e-05,
"loss": 2.1334,
"step": 810
},
{
"epoch": 0.71,
"learning_rate": 8.946778459004068e-05,
"loss": 2.1313,
"step": 815
},
{
"epoch": 0.72,
"learning_rate": 8.932135508378972e-05,
"loss": 2.1222,
"step": 820
},
{
"epoch": 0.72,
"learning_rate": 8.917403627350106e-05,
"loss": 2.1448,
"step": 825
},
{
"epoch": 0.73,
"learning_rate": 8.902583149098289e-05,
"loss": 2.141,
"step": 830
},
{
"epoch": 0.73,
"learning_rate": 8.887674408808082e-05,
"loss": 2.166,
"step": 835
},
{
"epoch": 0.74,
"learning_rate": 8.872677743660209e-05,
"loss": 2.1459,
"step": 840
},
{
"epoch": 0.74,
"learning_rate": 8.857593492823929e-05,
"loss": 2.1573,
"step": 845
},
{
"epoch": 0.74,
"learning_rate": 8.842421997449366e-05,
"loss": 2.1362,
"step": 850
},
{
"epoch": 0.75,
"learning_rate": 8.827163600659791e-05,
"loss": 2.1476,
"step": 855
},
{
"epoch": 0.75,
"learning_rate": 8.811818647543874e-05,
"loss": 2.1424,
"step": 860
},
{
"epoch": 0.76,
"learning_rate": 8.796387485147857e-05,
"loss": 2.1579,
"step": 865
},
{
"epoch": 0.76,
"learning_rate": 8.780870462467729e-05,
"loss": 2.1499,
"step": 870
},
{
"epoch": 0.77,
"learning_rate": 8.765267930441314e-05,
"loss": 2.1309,
"step": 875
},
{
"epoch": 0.77,
"learning_rate": 8.749580241940347e-05,
"loss": 2.1447,
"step": 880
},
{
"epoch": 0.78,
"learning_rate": 8.733807751762486e-05,
"loss": 2.1302,
"step": 885
},
{
"epoch": 0.78,
"learning_rate": 8.717950816623293e-05,
"loss": 2.1313,
"step": 890
},
{
"epoch": 0.78,
"learning_rate": 8.702009795148159e-05,
"loss": 2.1356,
"step": 895
},
{
"epoch": 0.79,
"learning_rate": 8.685985047864204e-05,
"loss": 2.1366,
"step": 900
},
{
"epoch": 0.79,
"eval_loss": 2.150692939758301,
"eval_runtime": 443.8324,
"eval_samples_per_second": 36.424,
"eval_steps_per_second": 1.14,
"step": 900
},
{
"epoch": 0.79,
"learning_rate": 8.66987693719211e-05,
"loss": 2.1463,
"step": 905
},
{
"epoch": 0.8,
"learning_rate": 8.653685827437936e-05,
"loss": 2.1394,
"step": 910
},
{
"epoch": 0.8,
"learning_rate": 8.637412084784873e-05,
"loss": 2.1548,
"step": 915
},
{
"epoch": 0.81,
"learning_rate": 8.621056077284964e-05,
"loss": 2.129,
"step": 920
},
{
"epoch": 0.81,
"learning_rate": 8.604618174850777e-05,
"loss": 2.1367,
"step": 925
},
{
"epoch": 0.81,
"learning_rate": 8.588098749247046e-05,
"loss": 2.1304,
"step": 930
},
{
"epoch": 0.82,
"learning_rate": 8.571498174082253e-05,
"loss": 2.1517,
"step": 935
},
{
"epoch": 0.82,
"learning_rate": 8.554816824800189e-05,
"loss": 2.1461,
"step": 940
},
{
"epoch": 0.83,
"learning_rate": 8.538055078671454e-05,
"loss": 2.144,
"step": 945
},
{
"epoch": 0.83,
"learning_rate": 8.521213314784931e-05,
"loss": 2.1571,
"step": 950
},
{
"epoch": 0.84,
"learning_rate": 8.504291914039204e-05,
"loss": 2.1504,
"step": 955
},
{
"epoch": 0.84,
"learning_rate": 8.487291259133956e-05,
"loss": 2.1348,
"step": 960
},
{
"epoch": 0.85,
"learning_rate": 8.470211734561305e-05,
"loss": 2.1241,
"step": 965
},
{
"epoch": 0.85,
"learning_rate": 8.453053726597103e-05,
"loss": 2.1386,
"step": 970
},
{
"epoch": 0.85,
"learning_rate": 8.43581762329222e-05,
"loss": 2.1515,
"step": 975
},
{
"epoch": 0.86,
"learning_rate": 8.418503814463741e-05,
"loss": 2.1459,
"step": 980
},
{
"epoch": 0.86,
"learning_rate": 8.401112691686175e-05,
"loss": 2.1281,
"step": 985
},
{
"epoch": 0.87,
"learning_rate": 8.383644648282578e-05,
"loss": 2.1318,
"step": 990
},
{
"epoch": 0.87,
"learning_rate": 8.366100079315672e-05,
"loss": 2.1453,
"step": 995
},
{
"epoch": 0.88,
"learning_rate": 8.348479381578904e-05,
"loss": 2.1219,
"step": 1000
},
{
"epoch": 0.88,
"eval_loss": 2.144993782043457,
"eval_runtime": 448.6393,
"eval_samples_per_second": 36.033,
"eval_steps_per_second": 1.128,
"step": 1000
},
{
"epoch": 0.88,
"learning_rate": 8.330782953587475e-05,
"loss": 2.1227,
"step": 1005
},
{
"epoch": 0.88,
"learning_rate": 8.31301119556932e-05,
"loss": 2.1257,
"step": 1010
},
{
"epoch": 0.89,
"learning_rate": 8.295164509456065e-05,
"loss": 2.1231,
"step": 1015
},
{
"epoch": 0.89,
"learning_rate": 8.277243298873936e-05,
"loss": 2.1199,
"step": 1020
},
{
"epoch": 0.9,
"learning_rate": 8.25924796913462e-05,
"loss": 2.1376,
"step": 1025
},
{
"epoch": 0.9,
"learning_rate": 8.241178927226112e-05,
"loss": 2.1166,
"step": 1030
},
{
"epoch": 0.91,
"learning_rate": 8.223036581803501e-05,
"loss": 2.1259,
"step": 1035
},
{
"epoch": 0.91,
"learning_rate": 8.204821343179737e-05,
"loss": 2.1397,
"step": 1040
},
{
"epoch": 0.92,
"learning_rate": 8.186533623316334e-05,
"loss": 2.138,
"step": 1045
},
{
"epoch": 0.92,
"learning_rate": 8.16817383581408e-05,
"loss": 2.1282,
"step": 1050
},
{
"epoch": 0.92,
"learning_rate": 8.149742395903653e-05,
"loss": 2.1212,
"step": 1055
},
{
"epoch": 0.93,
"learning_rate": 8.131239720436252e-05,
"loss": 2.1642,
"step": 1060
},
{
"epoch": 0.93,
"learning_rate": 8.11266622787416e-05,
"loss": 2.1365,
"step": 1065
},
{
"epoch": 0.94,
"learning_rate": 8.09402233828128e-05,
"loss": 2.127,
"step": 1070
},
{
"epoch": 0.94,
"learning_rate": 8.075308473313636e-05,
"loss": 2.1207,
"step": 1075
},
{
"epoch": 0.95,
"learning_rate": 8.056525056209841e-05,
"loss": 2.131,
"step": 1080
},
{
"epoch": 0.95,
"learning_rate": 8.037672511781516e-05,
"loss": 2.1154,
"step": 1085
},
{
"epoch": 0.95,
"learning_rate": 8.018751266403691e-05,
"loss": 2.1133,
"step": 1090
},
{
"epoch": 0.96,
"learning_rate": 7.999761748005156e-05,
"loss": 2.1324,
"step": 1095
},
{
"epoch": 0.96,
"learning_rate": 7.980704386058783e-05,
"loss": 2.1415,
"step": 1100
},
{
"epoch": 0.96,
"eval_loss": 2.1387038230895996,
"eval_runtime": 444.7162,
"eval_samples_per_second": 36.351,
"eval_steps_per_second": 1.138,
"step": 1100
},
{
"epoch": 0.97,
"learning_rate": 7.961579611571819e-05,
"loss": 2.1378,
"step": 1105
},
{
"epoch": 0.97,
"learning_rate": 7.94238785707613e-05,
"loss": 2.1092,
"step": 1110
},
{
"epoch": 0.98,
"learning_rate": 7.923129556618428e-05,
"loss": 2.103,
"step": 1115
},
{
"epoch": 0.98,
"learning_rate": 7.903805145750445e-05,
"loss": 2.1229,
"step": 1120
},
{
"epoch": 0.99,
"learning_rate": 7.884415061519086e-05,
"loss": 2.1163,
"step": 1125
},
{
"epoch": 0.99,
"learning_rate": 7.864959742456549e-05,
"loss": 2.1291,
"step": 1130
},
{
"epoch": 0.99,
"learning_rate": 7.845439628570402e-05,
"loss": 2.1214,
"step": 1135
},
{
"epoch": 1.0,
"learning_rate": 7.82585516133363e-05,
"loss": 2.1326,
"step": 1140
},
{
"epoch": 1.0,
"learning_rate": 7.806206783674657e-05,
"loss": 2.0425,
"step": 1145
},
{
"epoch": 1.01,
"learning_rate": 7.786494939967324e-05,
"loss": 1.9622,
"step": 1150
},
{
"epoch": 1.01,
"learning_rate": 7.76672007602084e-05,
"loss": 1.9902,
"step": 1155
},
{
"epoch": 1.02,
"learning_rate": 7.746882639069699e-05,
"loss": 1.999,
"step": 1160
},
{
"epoch": 1.02,
"learning_rate": 7.726983077763566e-05,
"loss": 1.979,
"step": 1165
},
{
"epoch": 1.02,
"learning_rate": 7.70702184215713e-05,
"loss": 1.9913,
"step": 1170
},
{
"epoch": 1.03,
"learning_rate": 7.686999383699924e-05,
"loss": 1.9861,
"step": 1175
},
{
"epoch": 1.03,
"learning_rate": 7.666916155226116e-05,
"loss": 1.9913,
"step": 1180
},
{
"epoch": 1.04,
"learning_rate": 7.646772610944271e-05,
"loss": 1.9892,
"step": 1185
},
{
"epoch": 1.04,
"learning_rate": 7.62656920642707e-05,
"loss": 1.9885,
"step": 1190
},
{
"epoch": 1.05,
"learning_rate": 7.606306398601017e-05,
"loss": 1.9982,
"step": 1195
},
{
"epoch": 1.05,
"learning_rate": 7.585984645736096e-05,
"loss": 1.9765,
"step": 1200
},
{
"epoch": 1.05,
"eval_loss": 2.144597053527832,
"eval_runtime": 441.7594,
"eval_samples_per_second": 36.595,
"eval_steps_per_second": 1.145,
"step": 1200
},
{
"epoch": 1.06,
"learning_rate": 7.56560440743541e-05,
"loss": 2.0011,
"step": 1205
},
{
"epoch": 1.06,
"learning_rate": 7.545166144624792e-05,
"loss": 1.9738,
"step": 1210
},
{
"epoch": 1.06,
"learning_rate": 7.52467031954237e-05,
"loss": 1.9889,
"step": 1215
},
{
"epoch": 1.07,
"learning_rate": 7.504117395728121e-05,
"loss": 1.986,
"step": 1220
},
{
"epoch": 1.07,
"learning_rate": 7.483507838013384e-05,
"loss": 1.9961,
"step": 1225
},
{
"epoch": 1.08,
"learning_rate": 7.462842112510346e-05,
"loss": 1.9755,
"step": 1230
},
{
"epoch": 1.08,
"learning_rate": 7.442120686601507e-05,
"loss": 1.9793,
"step": 1235
},
{
"epoch": 1.09,
"learning_rate": 7.421344028929102e-05,
"loss": 1.986,
"step": 1240
},
{
"epoch": 1.09,
"learning_rate": 7.4005126093845e-05,
"loss": 1.9839,
"step": 1245
},
{
"epoch": 1.1,
"learning_rate": 7.379626899097592e-05,
"loss": 1.9917,
"step": 1250
},
{
"epoch": 1.1,
"learning_rate": 7.358687370426117e-05,
"loss": 2.0014,
"step": 1255
},
{
"epoch": 1.1,
"learning_rate": 7.33769449694499e-05,
"loss": 1.9931,
"step": 1260
},
{
"epoch": 1.11,
"learning_rate": 7.31664875343559e-05,
"loss": 1.9819,
"step": 1265
},
{
"epoch": 1.11,
"learning_rate": 7.29555061587502e-05,
"loss": 1.9744,
"step": 1270
},
{
"epoch": 1.12,
"learning_rate": 7.274400561425341e-05,
"loss": 1.9886,
"step": 1275
},
{
"epoch": 1.12,
"learning_rate": 7.253199068422789e-05,
"loss": 1.9983,
"step": 1280
},
{
"epoch": 1.13,
"learning_rate": 7.231946616366943e-05,
"loss": 1.9928,
"step": 1285
},
{
"epoch": 1.13,
"learning_rate": 7.210643685909891e-05,
"loss": 1.9915,
"step": 1290
},
{
"epoch": 1.13,
"learning_rate": 7.189290758845355e-05,
"loss": 1.9745,
"step": 1295
},
{
"epoch": 1.14,
"learning_rate": 7.167888318097802e-05,
"loss": 1.9837,
"step": 1300
},
{
"epoch": 1.14,
"eval_loss": 2.1429882049560547,
"eval_runtime": 442.1904,
"eval_samples_per_second": 36.559,
"eval_steps_per_second": 1.144,
"step": 1300
},
{
"epoch": 1.14,
"learning_rate": 7.146436847711506e-05,
"loss": 1.9898,
"step": 1305
},
{
"epoch": 1.15,
"learning_rate": 7.124936832839621e-05,
"loss": 1.9762,
"step": 1310
},
{
"epoch": 1.15,
"learning_rate": 7.103388759733188e-05,
"loss": 1.9928,
"step": 1315
},
{
"epoch": 1.16,
"learning_rate": 7.081793115730153e-05,
"loss": 1.9774,
"step": 1320
},
{
"epoch": 1.16,
"learning_rate": 7.060150389244343e-05,
"loss": 1.9905,
"step": 1325
},
{
"epoch": 1.17,
"learning_rate": 7.038461069754414e-05,
"loss": 1.9779,
"step": 1330
},
{
"epoch": 1.17,
"learning_rate": 7.016725647792782e-05,
"loss": 1.9917,
"step": 1335
},
{
"epoch": 1.17,
"learning_rate": 6.994944614934535e-05,
"loss": 1.9901,
"step": 1340
},
{
"epoch": 1.18,
"learning_rate": 6.97311846378631e-05,
"loss": 2.011,
"step": 1345
},
{
"epoch": 1.18,
"learning_rate": 6.951247687975152e-05,
"loss": 1.9745,
"step": 1350
},
{
"epoch": 1.19,
"learning_rate": 6.929332782137348e-05,
"loss": 2.0014,
"step": 1355
},
{
"epoch": 1.19,
"learning_rate": 6.907374241907256e-05,
"loss": 1.9771,
"step": 1360
},
{
"epoch": 1.2,
"learning_rate": 6.885372563906067e-05,
"loss": 1.9963,
"step": 1365
},
{
"epoch": 1.2,
"learning_rate": 6.863328245730598e-05,
"loss": 1.9887,
"step": 1370
},
{
"epoch": 1.2,
"learning_rate": 6.841241785942032e-05,
"loss": 1.9948,
"step": 1375
},
{
"epoch": 1.21,
"learning_rate": 6.819113684054634e-05,
"loss": 1.9912,
"step": 1380
},
{
"epoch": 1.21,
"learning_rate": 6.79694444052446e-05,
"loss": 1.9855,
"step": 1385
},
{
"epoch": 1.22,
"learning_rate": 6.774734556738044e-05,
"loss": 1.9882,
"step": 1390
},
{
"epoch": 1.22,
"learning_rate": 6.752484535001045e-05,
"loss": 1.9877,
"step": 1395
},
{
"epoch": 1.23,
"learning_rate": 6.7301948785269e-05,
"loss": 1.9952,
"step": 1400
},
{
"epoch": 1.23,
"eval_loss": 2.1388282775878906,
"eval_runtime": 440.5731,
"eval_samples_per_second": 36.693,
"eval_steps_per_second": 1.149,
"step": 1400
},
{
"epoch": 1.23,
"learning_rate": 6.707866091425437e-05,
"loss": 1.9885,
"step": 1405
},
{
"epoch": 1.24,
"learning_rate": 6.685498678691471e-05,
"loss": 2.0204,
"step": 1410
},
{
"epoch": 1.24,
"learning_rate": 6.663093146193389e-05,
"loss": 1.9935,
"step": 1415
},
{
"epoch": 1.24,
"learning_rate": 6.640650000661708e-05,
"loss": 2.0011,
"step": 1420
},
{
"epoch": 1.25,
"learning_rate": 6.618169749677608e-05,
"loss": 1.9993,
"step": 1425
},
{
"epoch": 1.25,
"learning_rate": 6.595652901661464e-05,
"loss": 1.9818,
"step": 1430
},
{
"epoch": 1.26,
"learning_rate": 6.573099965861337e-05,
"loss": 1.9964,
"step": 1435
},
{
"epoch": 1.26,
"learning_rate": 6.550511452341459e-05,
"loss": 1.9809,
"step": 1440
},
{
"epoch": 1.27,
"learning_rate": 6.527887871970703e-05,
"loss": 1.992,
"step": 1445
},
{
"epoch": 1.27,
"learning_rate": 6.505229736411022e-05,
"loss": 1.9682,
"step": 1450
},
{
"epoch": 1.27,
"learning_rate": 6.482537558105885e-05,
"loss": 1.9893,
"step": 1455
},
{
"epoch": 1.28,
"learning_rate": 6.459811850268672e-05,
"loss": 1.996,
"step": 1460
},
{
"epoch": 1.28,
"learning_rate": 6.437053126871085e-05,
"loss": 1.9839,
"step": 1465
},
{
"epoch": 1.29,
"learning_rate": 6.414261902631515e-05,
"loss": 1.9839,
"step": 1470
},
{
"epoch": 1.29,
"learning_rate": 6.391438693003403e-05,
"loss": 1.9891,
"step": 1475
},
{
"epoch": 1.3,
"learning_rate": 6.368584014163581e-05,
"loss": 1.9768,
"step": 1480
},
{
"epoch": 1.3,
"learning_rate": 6.345698383000596e-05,
"loss": 1.971,
"step": 1485
},
{
"epoch": 1.31,
"learning_rate": 6.322782317103026e-05,
"loss": 1.9826,
"step": 1490
},
{
"epoch": 1.31,
"learning_rate": 6.299836334747768e-05,
"loss": 1.9897,
"step": 1495
},
{
"epoch": 1.31,
"learning_rate": 6.276860954888322e-05,
"loss": 1.9868,
"step": 1500
},
{
"epoch": 1.31,
"eval_loss": 2.1350769996643066,
"eval_runtime": 441.8563,
"eval_samples_per_second": 36.587,
"eval_steps_per_second": 1.145,
"step": 1500
},
{
"epoch": 1.32,
"learning_rate": 6.253856697143043e-05,
"loss": 1.9757,
"step": 1505
},
{
"epoch": 1.32,
"learning_rate": 6.23082408178341e-05,
"loss": 1.9987,
"step": 1510
},
{
"epoch": 1.33,
"learning_rate": 6.207763629722236e-05,
"loss": 2.0011,
"step": 1515
},
{
"epoch": 1.33,
"learning_rate": 6.184675862501901e-05,
"loss": 2.0104,
"step": 1520
},
{
"epoch": 1.34,
"learning_rate": 6.161561302282555e-05,
"loss": 1.9858,
"step": 1525
},
{
"epoch": 1.34,
"learning_rate": 6.138420471830307e-05,
"loss": 1.9906,
"step": 1530
},
{
"epoch": 1.34,
"learning_rate": 6.115253894505396e-05,
"loss": 1.9911,
"step": 1535
},
{
"epoch": 1.35,
"learning_rate": 6.092062094250367e-05,
"loss": 1.9828,
"step": 1540
},
{
"epoch": 1.35,
"learning_rate": 6.068845595578212e-05,
"loss": 1.9726,
"step": 1545
},
{
"epoch": 1.36,
"learning_rate": 6.0456049235605093e-05,
"loss": 1.985,
"step": 1550
},
{
"epoch": 1.36,
"learning_rate": 6.0223406038155514e-05,
"loss": 1.9846,
"step": 1555
},
{
"epoch": 1.37,
"learning_rate": 5.999053162496453e-05,
"loss": 2.0014,
"step": 1560
},
{
"epoch": 1.37,
"learning_rate": 5.975743126279254e-05,
"loss": 1.9893,
"step": 1565
},
{
"epoch": 1.38,
"learning_rate": 5.9524110223510076e-05,
"loss": 1.9907,
"step": 1570
},
{
"epoch": 1.38,
"learning_rate": 5.9290573783978585e-05,
"loss": 1.9987,
"step": 1575
},
{
"epoch": 1.38,
"learning_rate": 5.9056827225931064e-05,
"loss": 1.9957,
"step": 1580
},
{
"epoch": 1.39,
"learning_rate": 5.882287583585262e-05,
"loss": 1.9827,
"step": 1585
},
{
"epoch": 1.39,
"learning_rate": 5.858872490486089e-05,
"loss": 1.9926,
"step": 1590
},
{
"epoch": 1.4,
"learning_rate": 5.835437972858641e-05,
"loss": 1.9924,
"step": 1595
},
{
"epoch": 1.4,
"learning_rate": 5.8119845607052815e-05,
"loss": 1.9864,
"step": 1600
},
{
"epoch": 1.4,
"eval_loss": 2.131560802459717,
"eval_runtime": 441.1354,
"eval_samples_per_second": 36.646,
"eval_steps_per_second": 1.147,
"step": 1600
},
{
"epoch": 1.41,
"learning_rate": 5.7885127844556975e-05,
"loss": 1.9917,
"step": 1605
},
{
"epoch": 1.41,
"learning_rate": 5.7650231749549074e-05,
"loss": 1.9837,
"step": 1610
},
{
"epoch": 1.41,
"learning_rate": 5.741516263451251e-05,
"loss": 1.9881,
"step": 1615
},
{
"epoch": 1.42,
"learning_rate": 5.717992581584373e-05,
"loss": 1.9762,
"step": 1620
},
{
"epoch": 1.42,
"learning_rate": 5.694452661373203e-05,
"loss": 1.9964,
"step": 1625
},
{
"epoch": 1.43,
"learning_rate": 5.670897035203924e-05,
"loss": 1.9896,
"step": 1630
},
{
"epoch": 1.43,
"learning_rate": 5.647326235817928e-05,
"loss": 1.9837,
"step": 1635
},
{
"epoch": 1.44,
"learning_rate": 5.6237407962997704e-05,
"loss": 1.9859,
"step": 1640
},
{
"epoch": 1.44,
"learning_rate": 5.6001412500651104e-05,
"loss": 1.987,
"step": 1645
},
{
"epoch": 1.45,
"learning_rate": 5.5765281308486514e-05,
"loss": 1.9873,
"step": 1650
},
{
"epoch": 1.45,
"learning_rate": 5.5529019726920675e-05,
"loss": 1.978,
"step": 1655
},
{
"epoch": 1.45,
"learning_rate": 5.5292633099319224e-05,
"loss": 1.9852,
"step": 1660
},
{
"epoch": 1.46,
"learning_rate": 5.505612677187592e-05,
"loss": 1.9987,
"step": 1665
},
{
"epoch": 1.46,
"learning_rate": 5.481950609349166e-05,
"loss": 1.9821,
"step": 1670
},
{
"epoch": 1.47,
"learning_rate": 5.458277641565357e-05,
"loss": 1.9828,
"step": 1675
},
{
"epoch": 1.47,
"learning_rate": 5.434594309231389e-05,
"loss": 1.9924,
"step": 1680
},
{
"epoch": 1.48,
"learning_rate": 5.410901147976898e-05,
"loss": 1.9699,
"step": 1685
},
{
"epoch": 1.48,
"learning_rate": 5.387198693653813e-05,
"loss": 1.9827,
"step": 1690
},
{
"epoch": 1.48,
"learning_rate": 5.3634874823242386e-05,
"loss": 1.9694,
"step": 1695
},
{
"epoch": 1.49,
"learning_rate": 5.3397680502483306e-05,
"loss": 1.987,
"step": 1700
},
{
"epoch": 1.49,
"eval_loss": 2.1262922286987305,
"eval_runtime": 438.0155,
"eval_samples_per_second": 36.907,
"eval_steps_per_second": 1.155,
"step": 1700
},
{
"epoch": 1.49,
"learning_rate": 5.316040933872165e-05,
"loss": 1.9838,
"step": 1705
},
{
"epoch": 1.5,
"learning_rate": 5.292306669815611e-05,
"loss": 1.9873,
"step": 1710
},
{
"epoch": 1.5,
"learning_rate": 5.268565794860193e-05,
"loss": 1.9737,
"step": 1715
},
{
"epoch": 1.51,
"learning_rate": 5.244818845936944e-05,
"loss": 1.997,
"step": 1720
},
{
"epoch": 1.51,
"learning_rate": 5.221066360114275e-05,
"loss": 1.9945,
"step": 1725
},
{
"epoch": 1.52,
"learning_rate": 5.197308874585818e-05,
"loss": 1.9926,
"step": 1730
},
{
"epoch": 1.52,
"learning_rate": 5.173546926658277e-05,
"loss": 1.9759,
"step": 1735
},
{
"epoch": 1.52,
"learning_rate": 5.1497810537392844e-05,
"loss": 1.9713,
"step": 1740
},
{
"epoch": 1.53,
"learning_rate": 5.1260117933252384e-05,
"loss": 1.9984,
"step": 1745
},
{
"epoch": 1.53,
"learning_rate": 5.1022396829891484e-05,
"loss": 1.972,
"step": 1750
},
{
"epoch": 1.54,
"learning_rate": 5.0784652603684824e-05,
"loss": 1.9857,
"step": 1755
},
{
"epoch": 1.54,
"learning_rate": 5.0546890631530006e-05,
"loss": 1.9922,
"step": 1760
},
{
"epoch": 1.55,
"learning_rate": 5.0309116290726e-05,
"loss": 1.9865,
"step": 1765
},
{
"epoch": 1.55,
"learning_rate": 5.007133495885149e-05,
"loss": 1.9732,
"step": 1770
},
{
"epoch": 1.55,
"learning_rate": 4.9833552013643294e-05,
"loss": 1.9873,
"step": 1775
},
{
"epoch": 1.56,
"learning_rate": 4.959577283287471e-05,
"loss": 1.9925,
"step": 1780
},
{
"epoch": 1.56,
"learning_rate": 4.935800279423386e-05,
"loss": 1.984,
"step": 1785
},
{
"epoch": 1.57,
"learning_rate": 4.91202472752022e-05,
"loss": 1.9769,
"step": 1790
},
{
"epoch": 1.57,
"learning_rate": 4.888251165293267e-05,
"loss": 1.984,
"step": 1795
},
{
"epoch": 1.58,
"learning_rate": 4.8644801304128374e-05,
"loss": 1.9678,
"step": 1800
},
{
"epoch": 1.58,
"eval_loss": 2.1230146884918213,
"eval_runtime": 438.1696,
"eval_samples_per_second": 36.894,
"eval_steps_per_second": 1.155,
"step": 1800
},
{
"epoch": 1.58,
"learning_rate": 4.840712160492067e-05,
"loss": 1.9793,
"step": 1805
},
{
"epoch": 1.59,
"learning_rate": 4.8169477930747874e-05,
"loss": 1.9652,
"step": 1810
},
{
"epoch": 1.59,
"learning_rate": 4.7931875656233435e-05,
"loss": 1.9659,
"step": 1815
},
{
"epoch": 1.59,
"learning_rate": 4.769432015506456e-05,
"loss": 1.9991,
"step": 1820
},
{
"epoch": 1.6,
"learning_rate": 4.7456816799870607e-05,
"loss": 1.9674,
"step": 1825
},
{
"epoch": 1.6,
"learning_rate": 4.721937096210156e-05,
"loss": 1.9692,
"step": 1830
},
{
"epoch": 1.61,
"learning_rate": 4.698198801190659e-05,
"loss": 1.9937,
"step": 1835
},
{
"epoch": 1.61,
"learning_rate": 4.67446733180126e-05,
"loss": 1.9792,
"step": 1840
},
{
"epoch": 1.62,
"learning_rate": 4.6507432247602715e-05,
"loss": 1.9883,
"step": 1845
},
{
"epoch": 1.62,
"learning_rate": 4.627027016619508e-05,
"loss": 1.9606,
"step": 1850
},
{
"epoch": 1.63,
"learning_rate": 4.603319243752132e-05,
"loss": 1.9699,
"step": 1855
},
{
"epoch": 1.63,
"learning_rate": 4.57962044234053e-05,
"loss": 1.9701,
"step": 1860
},
{
"epoch": 1.63,
"learning_rate": 4.555931148364195e-05,
"loss": 1.9763,
"step": 1865
},
{
"epoch": 1.64,
"learning_rate": 4.5322518975875875e-05,
"loss": 1.978,
"step": 1870
},
{
"epoch": 1.64,
"learning_rate": 4.5085832255480386e-05,
"loss": 1.9652,
"step": 1875
},
{
"epoch": 1.65,
"learning_rate": 4.4849256675436165e-05,
"loss": 1.9806,
"step": 1880
},
{
"epoch": 1.65,
"learning_rate": 4.4612797586210394e-05,
"loss": 1.9621,
"step": 1885
},
{
"epoch": 1.66,
"learning_rate": 4.4376460335635594e-05,
"loss": 1.9786,
"step": 1890
},
{
"epoch": 1.66,
"learning_rate": 4.414025026878882e-05,
"loss": 1.9622,
"step": 1895
},
{
"epoch": 1.66,
"learning_rate": 4.3904172727870606e-05,
"loss": 1.9827,
"step": 1900
},
{
"epoch": 1.66,
"eval_loss": 2.1164419651031494,
"eval_runtime": 440.4446,
"eval_samples_per_second": 36.704,
"eval_steps_per_second": 1.149,
"step": 1900
},
{
"epoch": 1.67,
"learning_rate": 4.366823305208435e-05,
"loss": 1.9846,
"step": 1905
},
{
"epoch": 1.67,
"learning_rate": 4.343243657751533e-05,
"loss": 1.9577,
"step": 1910
},
{
"epoch": 1.68,
"learning_rate": 4.3196788637010254e-05,
"loss": 1.9577,
"step": 1915
},
{
"epoch": 1.68,
"learning_rate": 4.2961294560056445e-05,
"loss": 1.9646,
"step": 1920
},
{
"epoch": 1.69,
"learning_rate": 4.272595967266146e-05,
"loss": 1.9788,
"step": 1925
},
{
"epoch": 1.69,
"learning_rate": 4.249078929723256e-05,
"loss": 1.9553,
"step": 1930
},
{
"epoch": 1.7,
"learning_rate": 4.225578875245632e-05,
"loss": 1.9815,
"step": 1935
},
{
"epoch": 1.7,
"learning_rate": 4.202096335317841e-05,
"loss": 1.9916,
"step": 1940
},
{
"epoch": 1.7,
"learning_rate": 4.1786318410283345e-05,
"loss": 1.9741,
"step": 1945
},
{
"epoch": 1.71,
"learning_rate": 4.1551859230574375e-05,
"loss": 1.964,
"step": 1950
},
{
"epoch": 1.71,
"learning_rate": 4.131759111665349e-05,
"loss": 1.9781,
"step": 1955
},
{
"epoch": 1.72,
"learning_rate": 4.108351936680145e-05,
"loss": 1.9615,
"step": 1960
},
{
"epoch": 1.72,
"learning_rate": 4.0849649274857994e-05,
"loss": 1.9783,
"step": 1965
},
{
"epoch": 1.73,
"learning_rate": 4.061598613010213e-05,
"loss": 1.9516,
"step": 1970
},
{
"epoch": 1.73,
"learning_rate": 4.0382535217132414e-05,
"loss": 1.9868,
"step": 1975
},
{
"epoch": 1.73,
"learning_rate": 4.01493018157476e-05,
"loss": 1.9674,
"step": 1980
},
{
"epoch": 1.74,
"learning_rate": 3.991629120082703e-05,
"loss": 1.9773,
"step": 1985
},
{
"epoch": 1.74,
"learning_rate": 3.968350864221154e-05,
"loss": 1.9787,
"step": 1990
},
{
"epoch": 1.75,
"learning_rate": 3.945095940458407e-05,
"loss": 1.9728,
"step": 1995
},
{
"epoch": 1.75,
"learning_rate": 3.921864874735079e-05,
"loss": 1.9846,
"step": 2000
},
{
"epoch": 1.75,
"eval_loss": 2.113431930541992,
"eval_runtime": 438.4756,
"eval_samples_per_second": 36.869,
"eval_steps_per_second": 1.154,
"step": 2000
},
{
"epoch": 1.76,
"learning_rate": 3.8986581924522007e-05,
"loss": 1.9738,
"step": 2005
},
{
"epoch": 1.76,
"learning_rate": 3.875476418459339e-05,
"loss": 1.9683,
"step": 2010
},
{
"epoch": 1.77,
"learning_rate": 3.8523200770427326e-05,
"loss": 1.9774,
"step": 2015
},
{
"epoch": 1.77,
"learning_rate": 3.829189691913421e-05,
"loss": 1.9757,
"step": 2020
},
{
"epoch": 1.77,
"learning_rate": 3.806085786195417e-05,
"loss": 1.9761,
"step": 2025
},
{
"epoch": 1.78,
"learning_rate": 3.7830088824138585e-05,
"loss": 1.9855,
"step": 2030
},
{
"epoch": 1.78,
"learning_rate": 3.759959502483207e-05,
"loss": 1.9749,
"step": 2035
},
{
"epoch": 1.79,
"learning_rate": 3.7369381676954284e-05,
"loss": 1.9515,
"step": 2040
},
{
"epoch": 1.79,
"learning_rate": 3.713945398708217e-05,
"loss": 1.965,
"step": 2045
},
{
"epoch": 1.8,
"learning_rate": 3.690981715533211e-05,
"loss": 1.9592,
"step": 2050
},
{
"epoch": 1.8,
"learning_rate": 3.668047637524237e-05,
"loss": 1.9767,
"step": 2055
},
{
"epoch": 1.8,
"learning_rate": 3.645143683365555e-05,
"loss": 1.9668,
"step": 2060
},
{
"epoch": 1.81,
"learning_rate": 3.622270371060147e-05,
"loss": 1.9478,
"step": 2065
},
{
"epoch": 1.81,
"learning_rate": 3.5994282179179784e-05,
"loss": 1.9671,
"step": 2070
},
{
"epoch": 1.82,
"learning_rate": 3.576617740544312e-05,
"loss": 1.9796,
"step": 2075
},
{
"epoch": 1.82,
"learning_rate": 3.5538394548280255e-05,
"loss": 1.98,
"step": 2080
},
{
"epoch": 1.83,
"learning_rate": 3.531093875929934e-05,
"loss": 1.9799,
"step": 2085
},
{
"epoch": 1.83,
"learning_rate": 3.5083815182711524e-05,
"loss": 1.9712,
"step": 2090
},
{
"epoch": 1.84,
"learning_rate": 3.485702895521444e-05,
"loss": 1.9687,
"step": 2095
},
{
"epoch": 1.84,
"learning_rate": 3.463058520587625e-05,
"loss": 1.9694,
"step": 2100
},
{
"epoch": 1.84,
"eval_loss": 2.1068317890167236,
"eval_runtime": 436.7843,
"eval_samples_per_second": 37.011,
"eval_steps_per_second": 1.158,
"step": 2100
},
{
"epoch": 1.84,
"learning_rate": 3.440448905601939e-05,
"loss": 1.971,
"step": 2105
},
{
"epoch": 1.85,
"learning_rate": 3.417874561910498e-05,
"loss": 1.9587,
"step": 2110
},
{
"epoch": 1.85,
"learning_rate": 3.395336000061702e-05,
"loss": 1.9869,
"step": 2115
},
{
"epoch": 1.86,
"learning_rate": 3.372833729794703e-05,
"loss": 1.9621,
"step": 2120
},
{
"epoch": 1.86,
"learning_rate": 3.350368260027862e-05,
"loss": 1.9603,
"step": 2125
},
{
"epoch": 1.87,
"learning_rate": 3.32794009884726e-05,
"loss": 1.9715,
"step": 2130
},
{
"epoch": 1.87,
"learning_rate": 3.305549753495186e-05,
"loss": 1.9782,
"step": 2135
},
{
"epoch": 1.87,
"learning_rate": 3.283197730358678e-05,
"loss": 1.9622,
"step": 2140
},
{
"epoch": 1.88,
"learning_rate": 3.260884534958068e-05,
"loss": 1.9574,
"step": 2145
},
{
"epoch": 1.88,
"learning_rate": 3.2386106719355444e-05,
"loss": 1.9677,
"step": 2150
},
{
"epoch": 1.89,
"learning_rate": 3.216376645043747e-05,
"loss": 1.9754,
"step": 2155
},
{
"epoch": 1.89,
"learning_rate": 3.194182957134365e-05,
"loss": 1.9524,
"step": 2160
},
{
"epoch": 1.9,
"learning_rate": 3.172030110146771e-05,
"loss": 1.9704,
"step": 2165
},
{
"epoch": 1.9,
"learning_rate": 3.149918605096666e-05,
"loss": 1.9553,
"step": 2170
},
{
"epoch": 1.91,
"learning_rate": 3.127848942064747e-05,
"loss": 1.9811,
"step": 2175
},
{
"epoch": 1.91,
"learning_rate": 3.1058216201854007e-05,
"loss": 1.9698,
"step": 2180
},
{
"epoch": 1.91,
"learning_rate": 3.083837137635416e-05,
"loss": 1.9662,
"step": 2185
},
{
"epoch": 1.92,
"learning_rate": 3.0618959916227056e-05,
"loss": 1.9589,
"step": 2190
},
{
"epoch": 1.92,
"learning_rate": 3.0399986783750806e-05,
"loss": 1.9703,
"step": 2195
},
{
"epoch": 1.93,
"learning_rate": 3.0181456931290064e-05,
"loss": 1.9429,
"step": 2200
},
{
"epoch": 1.93,
"eval_loss": 2.10353946685791,
"eval_runtime": 434.9488,
"eval_samples_per_second": 37.168,
"eval_steps_per_second": 1.163,
"step": 2200
},
{
"epoch": 1.93,
"learning_rate": 2.9963375301184204e-05,
"loss": 1.9823,
"step": 2205
},
{
"epoch": 1.94,
"learning_rate": 2.9745746825635402e-05,
"loss": 1.9587,
"step": 2210
},
{
"epoch": 1.94,
"learning_rate": 2.952857642659721e-05,
"loss": 1.9622,
"step": 2215
},
{
"epoch": 1.94,
"learning_rate": 2.9311869015663125e-05,
"loss": 1.9736,
"step": 2220
},
{
"epoch": 1.95,
"learning_rate": 2.9095629493955573e-05,
"loss": 1.9581,
"step": 2225
},
{
"epoch": 1.95,
"learning_rate": 2.8879862752015075e-05,
"loss": 1.9603,
"step": 2230
},
{
"epoch": 1.96,
"learning_rate": 2.866457366968958e-05,
"loss": 1.96,
"step": 2235
},
{
"epoch": 1.96,
"learning_rate": 2.8449767116024194e-05,
"loss": 1.9828,
"step": 2240
},
{
"epoch": 1.97,
"learning_rate": 2.8235447949150927e-05,
"loss": 1.9684,
"step": 2245
},
{
"epoch": 1.97,
"learning_rate": 2.802162101617899e-05,
"loss": 1.9616,
"step": 2250
},
{
"epoch": 1.98,
"learning_rate": 2.7808291153084988e-05,
"loss": 1.9793,
"step": 2255
},
{
"epoch": 1.98,
"learning_rate": 2.7595463184603705e-05,
"loss": 1.9622,
"step": 2260
},
{
"epoch": 1.98,
"learning_rate": 2.7383141924118865e-05,
"loss": 1.9736,
"step": 2265
},
{
"epoch": 1.99,
"learning_rate": 2.7171332173554394e-05,
"loss": 1.9677,
"step": 2270
},
{
"epoch": 1.99,
"learning_rate": 2.696003872326567e-05,
"loss": 1.9506,
"step": 2275
},
{
"epoch": 2.0,
"learning_rate": 2.674926635193134e-05,
"loss": 1.9534,
"step": 2280
},
{
"epoch": 2.0,
"learning_rate": 2.6539019826445117e-05,
"loss": 1.8941,
"step": 2285
},
{
"epoch": 2.01,
"learning_rate": 2.632930390180801e-05,
"loss": 1.8062,
"step": 2290
},
{
"epoch": 2.01,
"learning_rate": 2.612012332102085e-05,
"loss": 1.796,
"step": 2295
},
{
"epoch": 2.01,
"learning_rate": 2.5911482814976912e-05,
"loss": 1.8079,
"step": 2300
},
{
"epoch": 2.01,
"eval_loss": 2.136899471282959,
"eval_runtime": 437.2683,
"eval_samples_per_second": 36.97,
"eval_steps_per_second": 1.157,
"step": 2300
},
{
"epoch": 2.02,
"learning_rate": 2.5703387102355014e-05,
"loss": 1.7992,
"step": 2305
},
{
"epoch": 2.02,
"learning_rate": 2.5495840889512697e-05,
"loss": 1.8124,
"step": 2310
},
{
"epoch": 2.03,
"learning_rate": 2.5288848870379922e-05,
"loss": 1.8099,
"step": 2315
},
{
"epoch": 2.03,
"learning_rate": 2.5082415726352748e-05,
"loss": 1.8173,
"step": 2320
},
{
"epoch": 2.04,
"learning_rate": 2.487654612618762e-05,
"loss": 1.8042,
"step": 2325
},
{
"epoch": 2.04,
"learning_rate": 2.4671244725895637e-05,
"loss": 1.8061,
"step": 2330
},
{
"epoch": 2.05,
"learning_rate": 2.446651616863736e-05,
"loss": 1.8031,
"step": 2335
},
{
"epoch": 2.05,
"learning_rate": 2.42623650846177e-05,
"loss": 1.8148,
"step": 2340
},
{
"epoch": 2.05,
"learning_rate": 2.4058796090981313e-05,
"loss": 1.8098,
"step": 2345
},
{
"epoch": 2.06,
"learning_rate": 2.385581379170806e-05,
"loss": 1.8066,
"step": 2350
},
{
"epoch": 2.06,
"learning_rate": 2.3653422777508994e-05,
"loss": 1.8125,
"step": 2355
},
{
"epoch": 2.07,
"learning_rate": 2.345162762572244e-05,
"loss": 1.8105,
"step": 2360
},
{
"epoch": 2.07,
"learning_rate": 2.3250432900210507e-05,
"loss": 1.8067,
"step": 2365
},
{
"epoch": 2.08,
"learning_rate": 2.3049843151255936e-05,
"loss": 1.8056,
"step": 2370
},
{
"epoch": 2.08,
"learning_rate": 2.2849862915459036e-05,
"loss": 1.7975,
"step": 2375
},
{
"epoch": 2.08,
"learning_rate": 2.2650496715635283e-05,
"loss": 1.8082,
"step": 2380
},
{
"epoch": 2.09,
"learning_rate": 2.245174906071282e-05,
"loss": 1.8125,
"step": 2385
},
{
"epoch": 2.09,
"learning_rate": 2.2253624445630673e-05,
"loss": 1.8078,
"step": 2390
},
{
"epoch": 2.1,
"learning_rate": 2.2056127351236934e-05,
"loss": 1.7967,
"step": 2395
},
{
"epoch": 2.1,
"learning_rate": 2.1859262244187556e-05,
"loss": 1.8132,
"step": 2400
},
{
"epoch": 2.1,
"eval_loss": 2.1375112533569336,
"eval_runtime": 437.2054,
"eval_samples_per_second": 36.976,
"eval_steps_per_second": 1.157,
"step": 2400
},
{
"epoch": 2.11,
"learning_rate": 2.1663033576845205e-05,
"loss": 1.8039,
"step": 2405
},
{
"epoch": 2.11,
"learning_rate": 2.1467445787178686e-05,
"loss": 1.7973,
"step": 2410
},
{
"epoch": 2.12,
"learning_rate": 2.1272503298662477e-05,
"loss": 1.8113,
"step": 2415
},
{
"epoch": 2.12,
"learning_rate": 2.1078210520176717e-05,
"loss": 1.8116,
"step": 2420
},
{
"epoch": 2.12,
"learning_rate": 2.0884571845907546e-05,
"loss": 1.809,
"step": 2425
},
{
"epoch": 2.13,
"learning_rate": 2.069159165524762e-05,
"loss": 1.8132,
"step": 2430
},
{
"epoch": 2.13,
"learning_rate": 2.04992743126972e-05,
"loss": 1.7959,
"step": 2435
},
{
"epoch": 2.14,
"learning_rate": 2.0307624167765288e-05,
"loss": 1.8103,
"step": 2440
},
{
"epoch": 2.14,
"learning_rate": 2.011664555487136e-05,
"loss": 1.8087,
"step": 2445
},
{
"epoch": 2.15,
"learning_rate": 1.992634279324736e-05,
"loss": 1.7934,
"step": 2450
},
{
"epoch": 2.15,
"learning_rate": 1.973672018683991e-05,
"loss": 1.8047,
"step": 2455
},
{
"epoch": 2.16,
"learning_rate": 1.9547782024213047e-05,
"loss": 1.7937,
"step": 2460
},
{
"epoch": 2.16,
"learning_rate": 1.9359532578451245e-05,
"loss": 1.8232,
"step": 2465
},
{
"epoch": 2.16,
"learning_rate": 1.9171976107062684e-05,
"loss": 1.8022,
"step": 2470
},
{
"epoch": 2.17,
"learning_rate": 1.8985116851883088e-05,
"loss": 1.8119,
"step": 2475
},
{
"epoch": 2.17,
"learning_rate": 1.8798959038979652e-05,
"loss": 1.7996,
"step": 2480
},
{
"epoch": 2.18,
"learning_rate": 1.8613506878555597e-05,
"loss": 1.8047,
"step": 2485
},
{
"epoch": 2.18,
"learning_rate": 1.8428764564854828e-05,
"loss": 1.7895,
"step": 2490
},
{
"epoch": 2.19,
"learning_rate": 1.8244736276067198e-05,
"loss": 1.8091,
"step": 2495
},
{
"epoch": 2.19,
"learning_rate": 1.80614261742339e-05,
"loss": 1.8043,
"step": 2500
},
{
"epoch": 2.19,
"eval_loss": 2.136007308959961,
"eval_runtime": 436.5151,
"eval_samples_per_second": 37.034,
"eval_steps_per_second": 1.159,
"step": 2500
},
{
"epoch": 2.19,
"learning_rate": 1.7878838405153385e-05,
"loss": 1.7965,
"step": 2505
},
{
"epoch": 2.2,
"learning_rate": 1.7696977098287654e-05,
"loss": 1.81,
"step": 2510
},
{
"epoch": 2.2,
"learning_rate": 1.7515846366668737e-05,
"loss": 1.8124,
"step": 2515
},
{
"epoch": 2.21,
"learning_rate": 1.7335450306805827e-05,
"loss": 1.7978,
"step": 2520
},
{
"epoch": 2.21,
"learning_rate": 1.715579299859248e-05,
"loss": 1.8076,
"step": 2525
},
{
"epoch": 2.22,
"learning_rate": 1.697687850521447e-05,
"loss": 1.8158,
"step": 2530
},
{
"epoch": 2.22,
"learning_rate": 1.6798710873057787e-05,
"loss": 1.7985,
"step": 2535
},
{
"epoch": 2.23,
"learning_rate": 1.662129413161723e-05,
"loss": 1.7988,
"step": 2540
},
{
"epoch": 2.23,
"learning_rate": 1.644463229340516e-05,
"loss": 1.7878,
"step": 2545
},
{
"epoch": 2.23,
"learning_rate": 1.6268729353860867e-05,
"loss": 1.8042,
"step": 2550
},
{
"epoch": 2.24,
"learning_rate": 1.6093589291260103e-05,
"loss": 1.8154,
"step": 2555
},
{
"epoch": 2.24,
"learning_rate": 1.5919216066625208e-05,
"loss": 1.7873,
"step": 2560
},
{
"epoch": 2.25,
"learning_rate": 1.5745613623635413e-05,
"loss": 1.8112,
"step": 2565
},
{
"epoch": 2.25,
"learning_rate": 1.557278588853779e-05,
"loss": 1.8059,
"step": 2570
},
{
"epoch": 2.26,
"learning_rate": 1.5400736770058294e-05,
"loss": 1.7951,
"step": 2575
},
{
"epoch": 2.26,
"learning_rate": 1.522947015931348e-05,
"loss": 1.807,
"step": 2580
},
{
"epoch": 2.26,
"learning_rate": 1.5058989929722495e-05,
"loss": 1.8121,
"step": 2585
},
{
"epoch": 2.27,
"learning_rate": 1.4889299936919371e-05,
"loss": 1.8003,
"step": 2590
},
{
"epoch": 2.27,
"learning_rate": 1.4720404018665984e-05,
"loss": 1.7861,
"step": 2595
},
{
"epoch": 2.28,
"learning_rate": 1.4552305994765097e-05,
"loss": 1.7927,
"step": 2600
},
{
"epoch": 2.28,
"eval_loss": 2.1333673000335693,
"eval_runtime": 436.639,
"eval_samples_per_second": 37.024,
"eval_steps_per_second": 1.159,
"step": 2600
},
{
"epoch": 2.28,
"learning_rate": 1.4385009666974097e-05,
"loss": 1.8122,
"step": 2605
},
{
"epoch": 2.29,
"learning_rate": 1.4218518818918925e-05,
"loss": 1.8045,
"step": 2610
},
{
"epoch": 2.29,
"learning_rate": 1.4052837216008585e-05,
"loss": 1.7979,
"step": 2615
},
{
"epoch": 2.3,
"learning_rate": 1.3887968605349877e-05,
"loss": 1.8117,
"step": 2620
},
{
"epoch": 2.3,
"learning_rate": 1.3723916715662782e-05,
"loss": 1.7819,
"step": 2625
},
{
"epoch": 2.3,
"learning_rate": 1.356068525719601e-05,
"loss": 1.807,
"step": 2630
},
{
"epoch": 2.31,
"learning_rate": 1.3398277921643188e-05,
"loss": 1.7996,
"step": 2635
},
{
"epoch": 2.31,
"learning_rate": 1.3236698382059287e-05,
"loss": 1.7984,
"step": 2640
},
{
"epoch": 2.32,
"learning_rate": 1.3075950292777588e-05,
"loss": 1.7975,
"step": 2645
},
{
"epoch": 2.32,
"learning_rate": 1.2916037289327065e-05,
"loss": 1.7987,
"step": 2650
},
{
"epoch": 2.33,
"learning_rate": 1.2756962988350108e-05,
"loss": 1.7958,
"step": 2655
},
{
"epoch": 2.33,
"learning_rate": 1.2598730987520718e-05,
"loss": 1.8064,
"step": 2660
},
{
"epoch": 2.33,
"learning_rate": 1.2441344865463245e-05,
"loss": 1.8101,
"step": 2665
},
{
"epoch": 2.34,
"learning_rate": 1.2284808181671325e-05,
"loss": 1.8195,
"step": 2670
},
{
"epoch": 2.34,
"learning_rate": 1.212912447642744e-05,
"loss": 1.7939,
"step": 2675
},
{
"epoch": 2.35,
"learning_rate": 1.1974297270722867e-05,
"loss": 1.8037,
"step": 2680
},
{
"epoch": 2.35,
"learning_rate": 1.1820330066177998e-05,
"loss": 1.7916,
"step": 2685
},
{
"epoch": 2.36,
"learning_rate": 1.1667226344963194e-05,
"loss": 1.7948,
"step": 2690
},
{
"epoch": 2.36,
"learning_rate": 1.151498956971997e-05,
"loss": 1.8054,
"step": 2695
},
{
"epoch": 2.37,
"learning_rate": 1.1363623183482775e-05,
"loss": 1.7935,
"step": 2700
},
{
"epoch": 2.37,
"eval_loss": 2.133516550064087,
"eval_runtime": 436.7537,
"eval_samples_per_second": 37.014,
"eval_steps_per_second": 1.159,
"step": 2700
},
{
"epoch": 2.37,
"learning_rate": 1.1213130609601003e-05,
"loss": 1.8049,
"step": 2705
},
{
"epoch": 2.37,
"learning_rate": 1.10635152516617e-05,
"loss": 1.8048,
"step": 2710
},
{
"epoch": 2.38,
"learning_rate": 1.0914780493412468e-05,
"loss": 1.812,
"step": 2715
},
{
"epoch": 2.38,
"learning_rate": 1.0766929698685041e-05,
"loss": 1.7978,
"step": 2720
},
{
"epoch": 2.39,
"learning_rate": 1.0619966211319115e-05,
"loss": 1.806,
"step": 2725
},
{
"epoch": 2.39,
"learning_rate": 1.0473893355086773e-05,
"loss": 1.803,
"step": 2730
},
{
"epoch": 2.4,
"learning_rate": 1.0328714433617348e-05,
"loss": 1.8001,
"step": 2735
},
{
"epoch": 2.4,
"learning_rate": 1.0184432730322601e-05,
"loss": 1.7951,
"step": 2740
},
{
"epoch": 2.4,
"learning_rate": 1.004105150832259e-05,
"loss": 1.8103,
"step": 2745
},
{
"epoch": 2.41,
"learning_rate": 9.898574010371747e-06,
"loss": 1.8028,
"step": 2750
},
{
"epoch": 2.41,
"learning_rate": 9.757003458785663e-06,
"loss": 1.8002,
"step": 2755
},
{
"epoch": 2.42,
"learning_rate": 9.616343055368083e-06,
"loss": 1.8008,
"step": 2760
},
{
"epoch": 2.42,
"learning_rate": 9.47659598133861e-06,
"loss": 1.8128,
"step": 2765
},
{
"epoch": 2.43,
"learning_rate": 9.337765397260656e-06,
"loss": 1.7967,
"step": 2770
},
{
"epoch": 2.43,
"learning_rate": 9.199854442970068e-06,
"loss": 1.8086,
"step": 2775
},
{
"epoch": 2.44,
"learning_rate": 9.062866237503993e-06,
"loss": 1.8022,
"step": 2780
},
{
"epoch": 2.44,
"learning_rate": 8.926803879030466e-06,
"loss": 1.8006,
"step": 2785
},
{
"epoch": 2.44,
"learning_rate": 8.791670444778228e-06,
"loss": 1.7894,
"step": 2790
},
{
"epoch": 2.45,
"learning_rate": 8.6574689909672e-06,
"loss": 1.798,
"step": 2795
},
{
"epoch": 2.45,
"learning_rate": 8.524202552739357e-06,
"loss": 1.7982,
"step": 2800
},
{
"epoch": 2.45,
"eval_loss": 2.1320557594299316,
"eval_runtime": 436.1068,
"eval_samples_per_second": 37.069,
"eval_steps_per_second": 1.16,
"step": 2800
},
{
"epoch": 2.46,
"learning_rate": 8.391874144090029e-06,
"loss": 1.794,
"step": 2805
},
{
"epoch": 2.46,
"learning_rate": 8.260486757799834e-06,
"loss": 1.8082,
"step": 2810
},
{
"epoch": 2.47,
"learning_rate": 8.130043365366873e-06,
"loss": 1.8078,
"step": 2815
},
{
"epoch": 2.47,
"learning_rate": 8.000546916939644e-06,
"loss": 1.8009,
"step": 2820
},
{
"epoch": 2.47,
"learning_rate": 7.872000341250213e-06,
"loss": 1.7924,
"step": 2825
},
{
"epoch": 2.48,
"learning_rate": 7.744406545548071e-06,
"loss": 1.8071,
"step": 2830
},
{
"epoch": 2.48,
"learning_rate": 7.617768415534305e-06,
"loss": 1.8094,
"step": 2835
},
{
"epoch": 2.49,
"learning_rate": 7.4920888152964e-06,
"loss": 1.8116,
"step": 2840
},
{
"epoch": 2.49,
"learning_rate": 7.367370587243394e-06,
"loss": 1.8,
"step": 2845
},
{
"epoch": 2.5,
"learning_rate": 7.243616552041677e-06,
"loss": 1.8054,
"step": 2850
},
{
"epoch": 2.5,
"learning_rate": 7.120829508551119e-06,
"loss": 1.8068,
"step": 2855
},
{
"epoch": 2.51,
"learning_rate": 6.999012233761798e-06,
"loss": 1.784,
"step": 2860
},
{
"epoch": 2.51,
"learning_rate": 6.878167482731251e-06,
"loss": 1.8162,
"step": 2865
},
{
"epoch": 2.51,
"learning_rate": 6.758297988522055e-06,
"loss": 1.799,
"step": 2870
},
{
"epoch": 2.52,
"learning_rate": 6.639406462140096e-06,
"loss": 1.7919,
"step": 2875
},
{
"epoch": 2.52,
"learning_rate": 6.521495592473259e-06,
"loss": 1.7979,
"step": 2880
},
{
"epoch": 2.53,
"learning_rate": 6.404568046230541e-06,
"loss": 1.8096,
"step": 2885
},
{
"epoch": 2.53,
"learning_rate": 6.2886264678818475e-06,
"loss": 1.812,
"step": 2890
},
{
"epoch": 2.54,
"learning_rate": 6.17367347959808e-06,
"loss": 1.8072,
"step": 2895
},
{
"epoch": 2.54,
"learning_rate": 6.059711681191888e-06,
"loss": 1.8029,
"step": 2900
},
{
"epoch": 2.54,
"eval_loss": 2.1311283111572266,
"eval_runtime": 435.6098,
"eval_samples_per_second": 37.111,
"eval_steps_per_second": 1.162,
"step": 2900
},
{
"epoch": 2.54,
"learning_rate": 5.946743650058889e-06,
"loss": 1.7884,
"step": 2905
},
{
"epoch": 2.55,
"learning_rate": 5.83477194111931e-06,
"loss": 1.796,
"step": 2910
},
{
"epoch": 2.55,
"learning_rate": 5.723799086760279e-06,
"loss": 1.8026,
"step": 2915
},
{
"epoch": 2.56,
"learning_rate": 5.613827596778487e-06,
"loss": 1.7973,
"step": 2920
},
{
"epoch": 2.56,
"learning_rate": 5.504859958323488e-06,
"loss": 1.8,
"step": 2925
},
{
"epoch": 2.57,
"learning_rate": 5.39689863584138e-06,
"loss": 1.7905,
"step": 2930
},
{
"epoch": 2.57,
"learning_rate": 5.289946071019142e-06,
"loss": 1.8092,
"step": 2935
},
{
"epoch": 2.58,
"learning_rate": 5.184004682729348e-06,
"loss": 1.7966,
"step": 2940
},
{
"epoch": 2.58,
"learning_rate": 5.0790768669754905e-06,
"loss": 1.7979,
"step": 2945
},
{
"epoch": 2.58,
"learning_rate": 4.975164996837811e-06,
"loss": 1.7885,
"step": 2950
},
{
"epoch": 2.59,
"learning_rate": 4.872271422419572e-06,
"loss": 1.7951,
"step": 2955
},
{
"epoch": 2.59,
"learning_rate": 4.770398470793991e-06,
"loss": 1.7922,
"step": 2960
},
{
"epoch": 2.6,
"learning_rate": 4.6695484459515045e-06,
"loss": 1.7961,
"step": 2965
},
{
"epoch": 2.6,
"learning_rate": 4.5697236287477685e-06,
"loss": 1.8079,
"step": 2970
},
{
"epoch": 2.61,
"learning_rate": 4.470926276851983e-06,
"loss": 1.7979,
"step": 2975
},
{
"epoch": 2.61,
"learning_rate": 4.373158624695917e-06,
"loss": 1.8001,
"step": 2980
},
{
"epoch": 2.61,
"learning_rate": 4.27642288342327e-06,
"loss": 1.8045,
"step": 2985
},
{
"epoch": 2.62,
"learning_rate": 4.18072124083978e-06,
"loss": 1.7914,
"step": 2990
},
{
"epoch": 2.62,
"learning_rate": 4.086055861363647e-06,
"loss": 1.7971,
"step": 2995
},
{
"epoch": 2.63,
"learning_rate": 3.992428885976652e-06,
"loss": 1.7919,
"step": 3000
},
{
"epoch": 2.63,
"eval_loss": 2.129803419113159,
"eval_runtime": 437.5268,
"eval_samples_per_second": 36.949,
"eval_steps_per_second": 1.157,
"step": 3000
},
{
"epoch": 2.63,
"learning_rate": 3.899842432175671e-06,
"loss": 1.8002,
"step": 3005
},
{
"epoch": 2.64,
"learning_rate": 3.8082985939248417e-06,
"loss": 1.7827,
"step": 3010
},
{
"epoch": 2.64,
"learning_rate": 3.7177994416081885e-06,
"loss": 1.7983,
"step": 3015
},
{
"epoch": 2.65,
"learning_rate": 3.6283470219827643e-06,
"loss": 1.7828,
"step": 3020
},
{
"epoch": 2.65,
"learning_rate": 3.539943358132425e-06,
"loss": 1.7992,
"step": 3025
},
{
"epoch": 2.65,
"learning_rate": 3.4525904494220106e-06,
"loss": 1.7907,
"step": 3030
},
{
"epoch": 2.66,
"learning_rate": 3.3662902714521714e-06,
"loss": 1.7855,
"step": 3035
},
{
"epoch": 2.66,
"learning_rate": 3.281044776014652e-06,
"loss": 1.7954,
"step": 3040
},
{
"epoch": 2.67,
"learning_rate": 3.196855891048195e-06,
"loss": 1.8015,
"step": 3045
},
{
"epoch": 2.67,
"learning_rate": 3.113725520594879e-06,
"loss": 1.7971,
"step": 3050
},
{
"epoch": 2.68,
"learning_rate": 3.0316555447571282e-06,
"loss": 1.8061,
"step": 3055
},
{
"epoch": 2.68,
"learning_rate": 2.9506478196551055e-06,
"loss": 1.7917,
"step": 3060
},
{
"epoch": 2.69,
"learning_rate": 2.87070417738482e-06,
"loss": 1.8035,
"step": 3065
},
{
"epoch": 2.69,
"learning_rate": 2.7918264259766126e-06,
"loss": 1.7921,
"step": 3070
},
{
"epoch": 2.69,
"learning_rate": 2.714016349354348e-06,
"loss": 1.7931,
"step": 3075
},
{
"epoch": 2.7,
"learning_rate": 2.6372757072949896e-06,
"loss": 1.7951,
"step": 3080
},
{
"epoch": 2.7,
"learning_rate": 2.5616062353888415e-06,
"loss": 1.8127,
"step": 3085
},
{
"epoch": 2.71,
"learning_rate": 2.48700964500031e-06,
"loss": 1.7914,
"step": 3090
},
{
"epoch": 2.71,
"learning_rate": 2.413487623229149e-06,
"loss": 1.797,
"step": 3095
},
{
"epoch": 2.72,
"learning_rate": 2.341041832872348e-06,
"loss": 1.7953,
"step": 3100
},
{
"epoch": 2.72,
"eval_loss": 2.1286771297454834,
"eval_runtime": 435.6868,
"eval_samples_per_second": 37.105,
"eval_steps_per_second": 1.161,
"step": 3100
},
{
"epoch": 2.72,
"learning_rate": 2.2696739123865218e-06,
"loss": 1.8038,
"step": 3105
},
{
"epoch": 2.72,
"learning_rate": 2.199385475850824e-06,
"loss": 1.7959,
"step": 3110
},
{
"epoch": 2.73,
"learning_rate": 2.1301781129304587e-06,
"loss": 1.8003,
"step": 3115
},
{
"epoch": 2.73,
"learning_rate": 2.062053388840768e-06,
"loss": 1.7858,
"step": 3120
},
{
"epoch": 2.74,
"learning_rate": 1.99501284431175e-06,
"loss": 1.7969,
"step": 3125
},
{
"epoch": 2.74,
"learning_rate": 1.929057995553296e-06,
"loss": 1.8083,
"step": 3130
},
{
"epoch": 2.75,
"learning_rate": 1.8641903342208388e-06,
"loss": 1.8103,
"step": 3135
},
{
"epoch": 2.75,
"learning_rate": 1.800411327381657e-06,
"loss": 1.7886,
"step": 3140
},
{
"epoch": 2.76,
"learning_rate": 1.7377224174816687e-06,
"loss": 1.7859,
"step": 3145
},
{
"epoch": 2.76,
"learning_rate": 1.6761250223128299e-06,
"loss": 1.8073,
"step": 3150
},
{
"epoch": 2.76,
"learning_rate": 1.6156205349810494e-06,
"loss": 1.8025,
"step": 3155
},
{
"epoch": 2.77,
"learning_rate": 1.556210323874685e-06,
"loss": 1.7855,
"step": 3160
},
{
"epoch": 2.77,
"learning_rate": 1.4978957326336252e-06,
"loss": 1.7863,
"step": 3165
},
{
"epoch": 2.78,
"learning_rate": 1.4406780801188514e-06,
"loss": 1.7958,
"step": 3170
},
{
"epoch": 2.78,
"learning_rate": 1.3845586603826677e-06,
"loss": 1.7935,
"step": 3175
},
{
"epoch": 2.79,
"learning_rate": 1.329538742639358e-06,
"loss": 1.7953,
"step": 3180
},
{
"epoch": 2.79,
"learning_rate": 1.2756195712365804e-06,
"loss": 1.791,
"step": 3185
},
{
"epoch": 2.79,
"learning_rate": 1.2228023656271182e-06,
"loss": 1.8091,
"step": 3190
},
{
"epoch": 2.8,
"learning_rate": 1.1710883203413904e-06,
"loss": 1.8014,
"step": 3195
},
{
"epoch": 2.8,
"learning_rate": 1.1204786049603789e-06,
"loss": 1.798,
"step": 3200
},
{
"epoch": 2.8,
"eval_loss": 2.127976179122925,
"eval_runtime": 437.1616,
"eval_samples_per_second": 36.979,
"eval_steps_per_second": 1.157,
"step": 3200
},
{
"epoch": 2.81,
"learning_rate": 1.0709743640892e-06,
"loss": 1.7822,
"step": 3205
},
{
"epoch": 2.81,
"learning_rate": 1.022576717331214e-06,
"loss": 1.7974,
"step": 3210
},
{
"epoch": 2.82,
"learning_rate": 9.752867592627112e-07,
"loss": 1.797,
"step": 3215
},
{
"epoch": 2.82,
"learning_rate": 9.291055594081278e-07,
"loss": 1.8001,
"step": 3220
},
{
"epoch": 2.83,
"learning_rate": 8.840341622159076e-07,
"loss": 1.7971,
"step": 3225
},
{
"epoch": 2.83,
"learning_rate": 8.400735870348231e-07,
"loss": 1.8057,
"step": 3230
},
{
"epoch": 2.83,
"learning_rate": 7.972248280909589e-07,
"loss": 1.7854,
"step": 3235
},
{
"epoch": 2.84,
"learning_rate": 7.554888544652305e-07,
"loss": 1.8117,
"step": 3240
},
{
"epoch": 2.84,
"learning_rate": 7.148666100714352e-07,
"loss": 1.779,
"step": 3245
},
{
"epoch": 2.85,
"learning_rate": 6.753590136349298e-07,
"loss": 1.8081,
"step": 3250
},
{
"epoch": 2.85,
"learning_rate": 6.369669586718419e-07,
"loss": 1.7938,
"step": 3255
},
{
"epoch": 2.86,
"learning_rate": 5.996913134688753e-07,
"loss": 1.8021,
"step": 3260
},
{
"epoch": 2.86,
"learning_rate": 5.635329210636531e-07,
"loss": 1.7888,
"step": 3265
},
{
"epoch": 2.86,
"learning_rate": 5.284925992256551e-07,
"loss": 1.7845,
"step": 3270
},
{
"epoch": 2.87,
"learning_rate": 4.945711404377385e-07,
"loss": 1.8015,
"step": 3275
},
{
"epoch": 2.87,
"learning_rate": 4.617693118782074e-07,
"loss": 1.7979,
"step": 3280
},
{
"epoch": 2.88,
"learning_rate": 4.3008785540344376e-07,
"loss": 1.7888,
"step": 3285
},
{
"epoch": 2.88,
"learning_rate": 3.9952748753115945e-07,
"loss": 1.7901,
"step": 3290
},
{
"epoch": 2.89,
"learning_rate": 3.7008889942417025e-07,
"loss": 1.7981,
"step": 3295
},
{
"epoch": 2.89,
"learning_rate": 3.4177275687476974e-07,
"loss": 1.7947,
"step": 3300
},
{
"epoch": 2.89,
"eval_loss": 2.1282496452331543,
"eval_runtime": 437.2091,
"eval_samples_per_second": 36.975,
"eval_steps_per_second": 1.157,
"step": 3300
},
{
"epoch": 2.9,
"learning_rate": 3.1457970028968e-07,
"loss": 1.7772,
"step": 3305
},
{
"epoch": 2.9,
"learning_rate": 2.885103446755466e-07,
"loss": 1.7805,
"step": 3310
},
{
"epoch": 2.9,
"learning_rate": 2.635652796250554e-07,
"loss": 1.7927,
"step": 3315
},
{
"epoch": 2.91,
"learning_rate": 2.3974506930358743e-07,
"loss": 1.7941,
"step": 3320
},
{
"epoch": 2.91,
"learning_rate": 2.1705025243644594e-07,
"loss": 1.7991,
"step": 3325
},
{
"epoch": 2.92,
"learning_rate": 1.9548134229669947e-07,
"loss": 1.7973,
"step": 3330
},
{
"epoch": 2.92,
"learning_rate": 1.7503882669354098e-07,
"loss": 1.802,
"step": 3335
},
{
"epoch": 2.93,
"learning_rate": 1.5572316796128583e-07,
"loss": 1.7945,
"step": 3340
},
{
"epoch": 2.93,
"learning_rate": 1.3753480294890763e-07,
"loss": 1.7975,
"step": 3345
},
{
"epoch": 2.93,
"learning_rate": 1.2047414301013527e-07,
"loss": 1.7829,
"step": 3350
},
{
"epoch": 2.94,
"learning_rate": 1.045415739941824e-07,
"loss": 1.8042,
"step": 3355
},
{
"epoch": 2.94,
"learning_rate": 8.973745623699903e-08,
"loss": 1.8032,
"step": 3360
},
{
"epoch": 2.95,
"learning_rate": 7.6062124553139e-08,
"loss": 1.8013,
"step": 3365
},
{
"epoch": 2.95,
"learning_rate": 6.351588822817167e-08,
"loss": 1.7983,
"step": 3370
},
{
"epoch": 2.96,
"learning_rate": 5.209903101168201e-08,
"loss": 1.7967,
"step": 3375
},
{
"epoch": 2.96,
"learning_rate": 4.1811811110886724e-08,
"loss": 1.7997,
"step": 3380
},
{
"epoch": 2.97,
"learning_rate": 3.265446118475568e-08,
"loss": 1.7907,
"step": 3385
},
{
"epoch": 2.97,
"learning_rate": 2.4627188338777196e-08,
"loss": 1.7963,
"step": 3390
},
{
"epoch": 2.97,
"learning_rate": 1.7730174120250687e-08,
"loss": 1.7791,
"step": 3395
},
{
"epoch": 2.98,
"learning_rate": 1.196357451421215e-08,
"loss": 1.8015,
"step": 3400
},
{
"epoch": 2.98,
"eval_loss": 2.1283345222473145,
"eval_runtime": 436.7591,
"eval_samples_per_second": 37.014,
"eval_steps_per_second": 1.159,
"step": 3400
},
{
"epoch": 2.98,
"learning_rate": 7.327519939870353e-09,
"loss": 1.7954,
"step": 3405
},
{
"epoch": 2.99,
"learning_rate": 3.822115247686941e-09,
"loss": 1.79,
"step": 3410
},
{
"epoch": 2.99,
"learning_rate": 1.4474397169894626e-09,
"loss": 1.79,
"step": 3415
},
{
"epoch": 3.0,
"learning_rate": 2.0354705417280351e-10,
"loss": 1.7978,
"step": 3420
},
{
"epoch": 3.0,
"step": 3423,
"total_flos": 5.633765648992567e+18,
"train_loss": 1.9906484538130407,
"train_runtime": 48039.6785,
"train_samples_per_second": 9.121,
"train_steps_per_second": 0.071
}
],
"logging_steps": 5,
"max_steps": 3423,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 100,
"total_flos": 5.633765648992567e+18,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}