Yi-6B-zhihu7 / trainer_state.json
yyx123's picture
Model save
11aae5b verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 20.0,
"eval_steps": 500,
"global_step": 1920,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 2.6041666666666667e-08,
"loss": 2.3378,
"step": 1
},
{
"epoch": 0.05,
"learning_rate": 1.3020833333333334e-07,
"loss": 2.6781,
"step": 5
},
{
"epoch": 0.1,
"learning_rate": 2.604166666666667e-07,
"loss": 2.6965,
"step": 10
},
{
"epoch": 0.16,
"learning_rate": 3.90625e-07,
"loss": 2.6689,
"step": 15
},
{
"epoch": 0.21,
"learning_rate": 5.208333333333334e-07,
"loss": 2.7272,
"step": 20
},
{
"epoch": 0.26,
"learning_rate": 6.510416666666668e-07,
"loss": 2.697,
"step": 25
},
{
"epoch": 0.31,
"learning_rate": 7.8125e-07,
"loss": 2.6381,
"step": 30
},
{
"epoch": 0.36,
"learning_rate": 9.114583333333333e-07,
"loss": 2.6254,
"step": 35
},
{
"epoch": 0.42,
"learning_rate": 1.0416666666666667e-06,
"loss": 2.6373,
"step": 40
},
{
"epoch": 0.47,
"learning_rate": 1.1718750000000001e-06,
"loss": 2.6103,
"step": 45
},
{
"epoch": 0.52,
"learning_rate": 1.3020833333333335e-06,
"loss": 2.5922,
"step": 50
},
{
"epoch": 0.57,
"learning_rate": 1.4322916666666667e-06,
"loss": 2.578,
"step": 55
},
{
"epoch": 0.62,
"learning_rate": 1.5625e-06,
"loss": 2.6198,
"step": 60
},
{
"epoch": 0.68,
"learning_rate": 1.6927083333333335e-06,
"loss": 2.6088,
"step": 65
},
{
"epoch": 0.73,
"learning_rate": 1.8229166666666666e-06,
"loss": 2.6466,
"step": 70
},
{
"epoch": 0.78,
"learning_rate": 1.953125e-06,
"loss": 2.6893,
"step": 75
},
{
"epoch": 0.83,
"learning_rate": 2.0833333333333334e-06,
"loss": 2.4503,
"step": 80
},
{
"epoch": 0.89,
"learning_rate": 2.213541666666667e-06,
"loss": 2.6438,
"step": 85
},
{
"epoch": 0.94,
"learning_rate": 2.3437500000000002e-06,
"loss": 2.6285,
"step": 90
},
{
"epoch": 0.99,
"learning_rate": 2.4739583333333336e-06,
"loss": 2.6831,
"step": 95
},
{
"epoch": 1.0,
"eval_loss": 2.6346399784088135,
"eval_runtime": 172.9512,
"eval_samples_per_second": 4.429,
"eval_steps_per_second": 1.11,
"step": 96
},
{
"epoch": 1.04,
"learning_rate": 2.604166666666667e-06,
"loss": 2.718,
"step": 100
},
{
"epoch": 1.09,
"learning_rate": 2.7343750000000004e-06,
"loss": 2.6197,
"step": 105
},
{
"epoch": 1.15,
"learning_rate": 2.8645833333333334e-06,
"loss": 2.6603,
"step": 110
},
{
"epoch": 1.2,
"learning_rate": 2.994791666666667e-06,
"loss": 2.7052,
"step": 115
},
{
"epoch": 1.25,
"learning_rate": 3.125e-06,
"loss": 2.6982,
"step": 120
},
{
"epoch": 1.3,
"learning_rate": 3.2552083333333335e-06,
"loss": 2.5888,
"step": 125
},
{
"epoch": 1.35,
"learning_rate": 3.385416666666667e-06,
"loss": 2.7218,
"step": 130
},
{
"epoch": 1.41,
"learning_rate": 3.5156250000000003e-06,
"loss": 2.6269,
"step": 135
},
{
"epoch": 1.46,
"learning_rate": 3.6458333333333333e-06,
"loss": 2.6213,
"step": 140
},
{
"epoch": 1.51,
"learning_rate": 3.776041666666667e-06,
"loss": 2.5227,
"step": 145
},
{
"epoch": 1.56,
"learning_rate": 3.90625e-06,
"loss": 2.6565,
"step": 150
},
{
"epoch": 1.61,
"learning_rate": 4.0364583333333335e-06,
"loss": 2.6106,
"step": 155
},
{
"epoch": 1.67,
"learning_rate": 4.166666666666667e-06,
"loss": 2.5478,
"step": 160
},
{
"epoch": 1.72,
"learning_rate": 4.296875e-06,
"loss": 2.6953,
"step": 165
},
{
"epoch": 1.77,
"learning_rate": 4.427083333333334e-06,
"loss": 2.7111,
"step": 170
},
{
"epoch": 1.82,
"learning_rate": 4.557291666666667e-06,
"loss": 2.661,
"step": 175
},
{
"epoch": 1.88,
"learning_rate": 4.6875000000000004e-06,
"loss": 2.5927,
"step": 180
},
{
"epoch": 1.93,
"learning_rate": 4.817708333333334e-06,
"loss": 2.4611,
"step": 185
},
{
"epoch": 1.98,
"learning_rate": 4.947916666666667e-06,
"loss": 2.6684,
"step": 190
},
{
"epoch": 2.0,
"eval_loss": 2.6288113594055176,
"eval_runtime": 172.9222,
"eval_samples_per_second": 4.43,
"eval_steps_per_second": 1.11,
"step": 192
},
{
"epoch": 2.03,
"learning_rate": 4.999962815348555e-06,
"loss": 2.5517,
"step": 195
},
{
"epoch": 2.08,
"learning_rate": 4.999735579817769e-06,
"loss": 2.6882,
"step": 200
},
{
"epoch": 2.14,
"learning_rate": 4.999301785650102e-06,
"loss": 2.6676,
"step": 205
},
{
"epoch": 2.19,
"learning_rate": 4.998661468690914e-06,
"loss": 2.5453,
"step": 210
},
{
"epoch": 2.24,
"learning_rate": 4.997814681851004e-06,
"loss": 2.7482,
"step": 215
},
{
"epoch": 2.29,
"learning_rate": 4.996761495102227e-06,
"loss": 2.6014,
"step": 220
},
{
"epoch": 2.34,
"learning_rate": 4.995501995471712e-06,
"loss": 2.6348,
"step": 225
},
{
"epoch": 2.4,
"learning_rate": 4.994036287034682e-06,
"loss": 2.6335,
"step": 230
},
{
"epoch": 2.45,
"learning_rate": 4.992364490905838e-06,
"loss": 2.5545,
"step": 235
},
{
"epoch": 2.5,
"learning_rate": 4.990486745229364e-06,
"loss": 2.5585,
"step": 240
},
{
"epoch": 2.55,
"learning_rate": 4.988403205167508e-06,
"loss": 2.6359,
"step": 245
},
{
"epoch": 2.6,
"learning_rate": 4.986114042887756e-06,
"loss": 2.5413,
"step": 250
},
{
"epoch": 2.66,
"learning_rate": 4.983619447548611e-06,
"loss": 2.6614,
"step": 255
},
{
"epoch": 2.71,
"learning_rate": 4.980919625283962e-06,
"loss": 2.6331,
"step": 260
},
{
"epoch": 2.76,
"learning_rate": 4.978014799186046e-06,
"loss": 2.6829,
"step": 265
},
{
"epoch": 2.81,
"learning_rate": 4.97490520928702e-06,
"loss": 2.6249,
"step": 270
},
{
"epoch": 2.86,
"learning_rate": 4.971591112539121e-06,
"loss": 2.6775,
"step": 275
},
{
"epoch": 2.92,
"learning_rate": 4.968072782793436e-06,
"loss": 2.5929,
"step": 280
},
{
"epoch": 2.97,
"learning_rate": 4.964350510777273e-06,
"loss": 2.687,
"step": 285
},
{
"epoch": 3.0,
"eval_loss": 2.6191797256469727,
"eval_runtime": 172.991,
"eval_samples_per_second": 4.428,
"eval_steps_per_second": 1.11,
"step": 288
},
{
"epoch": 3.02,
"learning_rate": 4.9604246040701384e-06,
"loss": 2.5492,
"step": 290
},
{
"epoch": 3.07,
"learning_rate": 4.956295387078319e-06,
"loss": 2.6643,
"step": 295
},
{
"epoch": 3.12,
"learning_rate": 4.9519632010080765e-06,
"loss": 2.6451,
"step": 300
},
{
"epoch": 3.18,
"learning_rate": 4.947428403837454e-06,
"loss": 2.447,
"step": 305
},
{
"epoch": 3.23,
"learning_rate": 4.942691370286693e-06,
"loss": 2.682,
"step": 310
},
{
"epoch": 3.28,
"learning_rate": 4.937752491787272e-06,
"loss": 2.5597,
"step": 315
},
{
"epoch": 3.33,
"learning_rate": 4.93261217644956e-06,
"loss": 2.5953,
"step": 320
},
{
"epoch": 3.39,
"learning_rate": 4.927270849029095e-06,
"loss": 2.5892,
"step": 325
},
{
"epoch": 3.44,
"learning_rate": 4.9217289508914836e-06,
"loss": 2.6764,
"step": 330
},
{
"epoch": 3.49,
"learning_rate": 4.91598693997593e-06,
"loss": 2.5029,
"step": 335
},
{
"epoch": 3.54,
"learning_rate": 4.910045290757399e-06,
"loss": 2.676,
"step": 340
},
{
"epoch": 3.59,
"learning_rate": 4.9039044942074055e-06,
"loss": 2.5671,
"step": 345
},
{
"epoch": 3.65,
"learning_rate": 4.8975650577534435e-06,
"loss": 2.6389,
"step": 350
},
{
"epoch": 3.7,
"learning_rate": 4.89102750523706e-06,
"loss": 2.5506,
"step": 355
},
{
"epoch": 3.75,
"learning_rate": 4.884292376870567e-06,
"loss": 2.5381,
"step": 360
},
{
"epoch": 3.8,
"learning_rate": 4.877360229192404e-06,
"loss": 2.6787,
"step": 365
},
{
"epoch": 3.85,
"learning_rate": 4.8702316350211445e-06,
"loss": 2.6745,
"step": 370
},
{
"epoch": 3.91,
"learning_rate": 4.86290718340817e-06,
"loss": 2.6619,
"step": 375
},
{
"epoch": 3.96,
"learning_rate": 4.855387479588991e-06,
"loss": 2.6597,
"step": 380
},
{
"epoch": 4.0,
"eval_loss": 2.6108741760253906,
"eval_runtime": 172.9898,
"eval_samples_per_second": 4.428,
"eval_steps_per_second": 1.11,
"step": 384
},
{
"epoch": 4.01,
"learning_rate": 4.847673144933239e-06,
"loss": 2.6435,
"step": 385
},
{
"epoch": 4.06,
"learning_rate": 4.839764816893315e-06,
"loss": 2.5638,
"step": 390
},
{
"epoch": 4.11,
"learning_rate": 4.83166314895172e-06,
"loss": 2.5875,
"step": 395
},
{
"epoch": 4.17,
"learning_rate": 4.823368810567056e-06,
"loss": 2.6563,
"step": 400
},
{
"epoch": 4.22,
"learning_rate": 4.814882487118708e-06,
"loss": 2.6723,
"step": 405
},
{
"epoch": 4.27,
"learning_rate": 4.806204879850206e-06,
"loss": 2.5723,
"step": 410
},
{
"epoch": 4.32,
"learning_rate": 4.797336705811282e-06,
"loss": 2.6688,
"step": 415
},
{
"epoch": 4.38,
"learning_rate": 4.788278697798619e-06,
"loss": 2.5098,
"step": 420
},
{
"epoch": 4.43,
"learning_rate": 4.779031604295297e-06,
"loss": 2.5736,
"step": 425
},
{
"epoch": 4.48,
"learning_rate": 4.769596189408947e-06,
"loss": 2.5369,
"step": 430
},
{
"epoch": 4.53,
"learning_rate": 4.759973232808609e-06,
"loss": 2.5826,
"step": 435
},
{
"epoch": 4.58,
"learning_rate": 4.750163529660303e-06,
"loss": 2.722,
"step": 440
},
{
"epoch": 4.64,
"learning_rate": 4.740167890561331e-06,
"loss": 2.5048,
"step": 445
},
{
"epoch": 4.69,
"learning_rate": 4.729987141473286e-06,
"loss": 2.7039,
"step": 450
},
{
"epoch": 4.74,
"learning_rate": 4.719622123653809e-06,
"loss": 2.5838,
"step": 455
},
{
"epoch": 4.79,
"learning_rate": 4.70907369358707e-06,
"loss": 2.5589,
"step": 460
},
{
"epoch": 4.84,
"learning_rate": 4.698342722912993e-06,
"loss": 2.6799,
"step": 465
},
{
"epoch": 4.9,
"learning_rate": 4.687430098355237e-06,
"loss": 2.654,
"step": 470
},
{
"epoch": 4.95,
"learning_rate": 4.676336721647917e-06,
"loss": 2.6297,
"step": 475
},
{
"epoch": 5.0,
"learning_rate": 4.665063509461098e-06,
"loss": 2.6019,
"step": 480
},
{
"epoch": 5.0,
"eval_loss": 2.6053690910339355,
"eval_runtime": 172.9519,
"eval_samples_per_second": 4.429,
"eval_steps_per_second": 1.11,
"step": 480
},
{
"epoch": 5.05,
"learning_rate": 4.6536113933250395e-06,
"loss": 2.6562,
"step": 485
},
{
"epoch": 5.1,
"learning_rate": 4.641981319553232e-06,
"loss": 2.654,
"step": 490
},
{
"epoch": 5.16,
"learning_rate": 4.630174249164197e-06,
"loss": 2.5684,
"step": 495
},
{
"epoch": 5.21,
"learning_rate": 4.61819115780207e-06,
"loss": 2.5459,
"step": 500
},
{
"epoch": 5.26,
"learning_rate": 4.606033035655989e-06,
"loss": 2.5952,
"step": 505
},
{
"epoch": 5.31,
"learning_rate": 4.59370088737827e-06,
"loss": 2.7195,
"step": 510
},
{
"epoch": 5.36,
"learning_rate": 4.581195732001388e-06,
"loss": 2.591,
"step": 515
},
{
"epoch": 5.42,
"learning_rate": 4.568518602853776e-06,
"loss": 2.5691,
"step": 520
},
{
"epoch": 5.47,
"learning_rate": 4.555670547474438e-06,
"loss": 2.5543,
"step": 525
},
{
"epoch": 5.52,
"learning_rate": 4.542652627526385e-06,
"loss": 2.589,
"step": 530
},
{
"epoch": 5.57,
"learning_rate": 4.529465918708911e-06,
"loss": 2.6588,
"step": 535
},
{
"epoch": 5.62,
"learning_rate": 4.516111510668707e-06,
"loss": 2.5623,
"step": 540
},
{
"epoch": 5.68,
"learning_rate": 4.502590506909815e-06,
"loss": 2.6214,
"step": 545
},
{
"epoch": 5.73,
"learning_rate": 4.48890402470245e-06,
"loss": 2.6125,
"step": 550
},
{
"epoch": 5.78,
"learning_rate": 4.475053194990673e-06,
"loss": 2.6154,
"step": 555
},
{
"epoch": 5.83,
"learning_rate": 4.46103916229894e-06,
"loss": 2.5869,
"step": 560
},
{
"epoch": 5.89,
"learning_rate": 4.446863084637527e-06,
"loss": 2.5409,
"step": 565
},
{
"epoch": 5.94,
"learning_rate": 4.432526133406843e-06,
"loss": 2.6322,
"step": 570
},
{
"epoch": 5.99,
"learning_rate": 4.4180294933006315e-06,
"loss": 2.6118,
"step": 575
},
{
"epoch": 6.0,
"eval_loss": 2.602187395095825,
"eval_runtime": 172.8706,
"eval_samples_per_second": 4.431,
"eval_steps_per_second": 1.111,
"step": 576
},
{
"epoch": 6.04,
"learning_rate": 4.403374362208078e-06,
"loss": 2.5985,
"step": 580
},
{
"epoch": 6.09,
"learning_rate": 4.388561951114829e-06,
"loss": 2.6755,
"step": 585
},
{
"epoch": 6.15,
"learning_rate": 4.3735934840029185e-06,
"loss": 2.5744,
"step": 590
},
{
"epoch": 6.2,
"learning_rate": 4.358470197749635e-06,
"loss": 2.6534,
"step": 595
},
{
"epoch": 6.25,
"learning_rate": 4.34319334202531e-06,
"loss": 2.4585,
"step": 600
},
{
"epoch": 6.3,
"learning_rate": 4.327764179190061e-06,
"loss": 2.5897,
"step": 605
},
{
"epoch": 6.35,
"learning_rate": 4.312183984189471e-06,
"loss": 2.6968,
"step": 610
},
{
"epoch": 6.41,
"learning_rate": 4.296454044449245e-06,
"loss": 2.598,
"step": 615
},
{
"epoch": 6.46,
"learning_rate": 4.280575659768828e-06,
"loss": 2.614,
"step": 620
},
{
"epoch": 6.51,
"learning_rate": 4.26455014221399e-06,
"loss": 2.5943,
"step": 625
},
{
"epoch": 6.56,
"learning_rate": 4.248378816008418e-06,
"loss": 2.6974,
"step": 630
},
{
"epoch": 6.61,
"learning_rate": 4.232063017424285e-06,
"loss": 2.5015,
"step": 635
},
{
"epoch": 6.67,
"learning_rate": 4.215604094671835e-06,
"loss": 2.5402,
"step": 640
},
{
"epoch": 6.72,
"learning_rate": 4.1990034077879746e-06,
"loss": 2.5983,
"step": 645
},
{
"epoch": 6.77,
"learning_rate": 4.1822623285238944e-06,
"loss": 2.5948,
"step": 650
},
{
"epoch": 6.82,
"learning_rate": 4.165382240231713e-06,
"loss": 2.5587,
"step": 655
},
{
"epoch": 6.88,
"learning_rate": 4.1483645377501726e-06,
"loss": 2.5544,
"step": 660
},
{
"epoch": 6.93,
"learning_rate": 4.1312106272893745e-06,
"loss": 2.5757,
"step": 665
},
{
"epoch": 6.98,
"learning_rate": 4.113921926314587e-06,
"loss": 2.7286,
"step": 670
},
{
"epoch": 7.0,
"eval_loss": 2.600128173828125,
"eval_runtime": 172.8748,
"eval_samples_per_second": 4.431,
"eval_steps_per_second": 1.111,
"step": 672
},
{
"epoch": 7.03,
"learning_rate": 4.096499863429113e-06,
"loss": 2.5867,
"step": 675
},
{
"epoch": 7.08,
"learning_rate": 4.078945878256244e-06,
"loss": 2.6628,
"step": 680
},
{
"epoch": 7.14,
"learning_rate": 4.061261421320298e-06,
"loss": 2.6499,
"step": 685
},
{
"epoch": 7.19,
"learning_rate": 4.043447953926763e-06,
"loss": 2.5979,
"step": 690
},
{
"epoch": 7.24,
"learning_rate": 4.025506948041542e-06,
"loss": 2.5506,
"step": 695
},
{
"epoch": 7.29,
"learning_rate": 4.0074398861693244e-06,
"loss": 2.5433,
"step": 700
},
{
"epoch": 7.34,
"learning_rate": 3.989248261231084e-06,
"loss": 2.5475,
"step": 705
},
{
"epoch": 7.4,
"learning_rate": 3.9709335764407115e-06,
"loss": 2.781,
"step": 710
},
{
"epoch": 7.45,
"learning_rate": 3.952497345180806e-06,
"loss": 2.556,
"step": 715
},
{
"epoch": 7.5,
"learning_rate": 3.933941090877615e-06,
"loss": 2.6076,
"step": 720
},
{
"epoch": 7.55,
"learning_rate": 3.915266346875157e-06,
"loss": 2.5666,
"step": 725
},
{
"epoch": 7.6,
"learning_rate": 3.896474656308512e-06,
"loss": 2.5663,
"step": 730
},
{
"epoch": 7.66,
"learning_rate": 3.877567571976311e-06,
"loss": 2.5254,
"step": 735
},
{
"epoch": 7.71,
"learning_rate": 3.858546656212425e-06,
"loss": 2.6294,
"step": 740
},
{
"epoch": 7.76,
"learning_rate": 3.8394134807568695e-06,
"loss": 2.5674,
"step": 745
},
{
"epoch": 7.81,
"learning_rate": 3.82016962662592e-06,
"loss": 2.6185,
"step": 750
},
{
"epoch": 7.86,
"learning_rate": 3.8008166839814776e-06,
"loss": 2.5857,
"step": 755
},
{
"epoch": 7.92,
"learning_rate": 3.7813562519996633e-06,
"loss": 2.6275,
"step": 760
},
{
"epoch": 7.97,
"learning_rate": 3.7617899387386803e-06,
"loss": 2.6341,
"step": 765
},
{
"epoch": 8.0,
"eval_loss": 2.5987157821655273,
"eval_runtime": 173.4207,
"eval_samples_per_second": 4.417,
"eval_steps_per_second": 1.107,
"step": 768
},
{
"epoch": 8.02,
"learning_rate": 3.7421193610059347e-06,
"loss": 2.6327,
"step": 770
},
{
"epoch": 8.07,
"learning_rate": 3.722346144224432e-06,
"loss": 2.5603,
"step": 775
},
{
"epoch": 8.12,
"learning_rate": 3.7024719222984696e-06,
"loss": 2.5164,
"step": 780
},
{
"epoch": 8.18,
"learning_rate": 3.6824983374786216e-06,
"loss": 2.6151,
"step": 785
},
{
"epoch": 8.23,
"learning_rate": 3.6624270402260355e-06,
"loss": 2.7189,
"step": 790
},
{
"epoch": 8.28,
"learning_rate": 3.642259689076052e-06,
"loss": 2.5992,
"step": 795
},
{
"epoch": 8.33,
"learning_rate": 3.621997950501156e-06,
"loss": 2.571,
"step": 800
},
{
"epoch": 8.39,
"learning_rate": 3.6016434987732716e-06,
"loss": 2.6104,
"step": 805
},
{
"epoch": 8.44,
"learning_rate": 3.5811980158254156e-06,
"loss": 2.5709,
"step": 810
},
{
"epoch": 8.49,
"learning_rate": 3.5606631911127143e-06,
"loss": 2.6242,
"step": 815
},
{
"epoch": 8.54,
"learning_rate": 3.5400407214727983e-06,
"loss": 2.5912,
"step": 820
},
{
"epoch": 8.59,
"learning_rate": 3.519332310985592e-06,
"loss": 2.65,
"step": 825
},
{
"epoch": 8.65,
"learning_rate": 3.4985396708325014e-06,
"loss": 2.5836,
"step": 830
},
{
"epoch": 8.7,
"learning_rate": 3.477664519155014e-06,
"loss": 2.5325,
"step": 835
},
{
"epoch": 8.75,
"learning_rate": 3.4567085809127247e-06,
"loss": 2.5664,
"step": 840
},
{
"epoch": 8.8,
"learning_rate": 3.435673587740801e-06,
"loss": 2.56,
"step": 845
},
{
"epoch": 8.85,
"learning_rate": 3.414561277806893e-06,
"loss": 2.6283,
"step": 850
},
{
"epoch": 8.91,
"learning_rate": 3.3933733956675006e-06,
"loss": 2.6294,
"step": 855
},
{
"epoch": 8.96,
"learning_rate": 3.3721116921238273e-06,
"loss": 2.572,
"step": 860
},
{
"epoch": 9.0,
"eval_loss": 2.59786319732666,
"eval_runtime": 173.4256,
"eval_samples_per_second": 4.417,
"eval_steps_per_second": 1.107,
"step": 864
},
{
"epoch": 9.01,
"learning_rate": 3.3507779240770967e-06,
"loss": 2.6574,
"step": 865
},
{
"epoch": 9.06,
"learning_rate": 3.3293738543833807e-06,
"loss": 2.6334,
"step": 870
},
{
"epoch": 9.11,
"learning_rate": 3.3079012517079325e-06,
"loss": 2.5623,
"step": 875
},
{
"epoch": 9.17,
"learning_rate": 3.2863618903790346e-06,
"loss": 2.566,
"step": 880
},
{
"epoch": 9.22,
"learning_rate": 3.264757550241384e-06,
"loss": 2.571,
"step": 885
},
{
"epoch": 9.27,
"learning_rate": 3.2430900165090195e-06,
"loss": 2.5241,
"step": 890
},
{
"epoch": 9.32,
"learning_rate": 3.221361079617804e-06,
"loss": 2.5838,
"step": 895
},
{
"epoch": 9.38,
"learning_rate": 3.199572535077481e-06,
"loss": 2.6498,
"step": 900
},
{
"epoch": 9.43,
"learning_rate": 3.177726183323301e-06,
"loss": 2.4745,
"step": 905
},
{
"epoch": 9.48,
"learning_rate": 3.1558238295672544e-06,
"loss": 2.6874,
"step": 910
},
{
"epoch": 9.53,
"learning_rate": 3.1338672836488994e-06,
"loss": 2.5929,
"step": 915
},
{
"epoch": 9.58,
"learning_rate": 3.1118583598858097e-06,
"loss": 2.5818,
"step": 920
},
{
"epoch": 9.64,
"learning_rate": 3.0897988769236583e-06,
"loss": 2.6143,
"step": 925
},
{
"epoch": 9.69,
"learning_rate": 3.0676906575859335e-06,
"loss": 2.5951,
"step": 930
},
{
"epoch": 9.74,
"learning_rate": 3.0455355287233175e-06,
"loss": 2.5665,
"step": 935
},
{
"epoch": 9.79,
"learning_rate": 3.0233353210627305e-06,
"loss": 2.6756,
"step": 940
},
{
"epoch": 9.84,
"learning_rate": 3.0010918690560516e-06,
"loss": 2.7057,
"step": 945
},
{
"epoch": 9.9,
"learning_rate": 2.978807010728536e-06,
"loss": 2.5657,
"step": 950
},
{
"epoch": 9.95,
"learning_rate": 2.9564825875269333e-06,
"loss": 2.574,
"step": 955
},
{
"epoch": 10.0,
"learning_rate": 2.9341204441673267e-06,
"loss": 2.622,
"step": 960
},
{
"epoch": 10.0,
"eval_loss": 2.5974395275115967,
"eval_runtime": 173.3905,
"eval_samples_per_second": 4.418,
"eval_steps_per_second": 1.107,
"step": 960
},
{
"epoch": 10.05,
"learning_rate": 2.9117224284826966e-06,
"loss": 2.6295,
"step": 965
},
{
"epoch": 10.1,
"learning_rate": 2.889290391270235e-06,
"loss": 2.7132,
"step": 970
},
{
"epoch": 10.16,
"learning_rate": 2.8668261861384045e-06,
"loss": 2.637,
"step": 975
},
{
"epoch": 10.21,
"learning_rate": 2.844331669353777e-06,
"loss": 2.6434,
"step": 980
},
{
"epoch": 10.26,
"learning_rate": 2.8218086996876405e-06,
"loss": 2.6185,
"step": 985
},
{
"epoch": 10.31,
"learning_rate": 2.7992591382624064e-06,
"loss": 2.6806,
"step": 990
},
{
"epoch": 10.36,
"learning_rate": 2.7766848483978235e-06,
"loss": 2.4173,
"step": 995
},
{
"epoch": 10.42,
"learning_rate": 2.754087695457005e-06,
"loss": 2.7256,
"step": 1000
},
{
"epoch": 10.47,
"learning_rate": 2.7314695466922896e-06,
"loss": 2.5551,
"step": 1005
},
{
"epoch": 10.52,
"learning_rate": 2.708832271090947e-06,
"loss": 2.5539,
"step": 1010
},
{
"epoch": 10.57,
"learning_rate": 2.6861777392207415e-06,
"loss": 2.5384,
"step": 1015
},
{
"epoch": 10.62,
"learning_rate": 2.663507823075358e-06,
"loss": 2.5276,
"step": 1020
},
{
"epoch": 10.68,
"learning_rate": 2.6408243959197192e-06,
"loss": 2.5189,
"step": 1025
},
{
"epoch": 10.73,
"learning_rate": 2.6181293321351925e-06,
"loss": 2.6471,
"step": 1030
},
{
"epoch": 10.78,
"learning_rate": 2.5954245070647042e-06,
"loss": 2.5109,
"step": 1035
},
{
"epoch": 10.83,
"learning_rate": 2.572711796857779e-06,
"loss": 2.5946,
"step": 1040
},
{
"epoch": 10.89,
"learning_rate": 2.549993078315505e-06,
"loss": 2.5543,
"step": 1045
},
{
"epoch": 10.94,
"learning_rate": 2.527270228735456e-06,
"loss": 2.6536,
"step": 1050
},
{
"epoch": 10.99,
"learning_rate": 2.5045451257565597e-06,
"loss": 2.6404,
"step": 1055
},
{
"epoch": 11.0,
"eval_loss": 2.597221851348877,
"eval_runtime": 173.4251,
"eval_samples_per_second": 4.417,
"eval_steps_per_second": 1.107,
"step": 1056
},
{
"epoch": 11.04,
"learning_rate": 2.4818196472039464e-06,
"loss": 2.5136,
"step": 1060
},
{
"epoch": 11.09,
"learning_rate": 2.459095670933783e-06,
"loss": 2.6783,
"step": 1065
},
{
"epoch": 11.15,
"learning_rate": 2.4363750746781e-06,
"loss": 2.656,
"step": 1070
},
{
"epoch": 11.2,
"learning_rate": 2.4136597358896273e-06,
"loss": 2.7115,
"step": 1075
},
{
"epoch": 11.25,
"learning_rate": 2.3909515315866606e-06,
"loss": 2.5209,
"step": 1080
},
{
"epoch": 11.3,
"learning_rate": 2.368252338197956e-06,
"loss": 2.5409,
"step": 1085
},
{
"epoch": 11.35,
"learning_rate": 2.3455640314076805e-06,
"loss": 2.62,
"step": 1090
},
{
"epoch": 11.41,
"learning_rate": 2.322888486000415e-06,
"loss": 2.6359,
"step": 1095
},
{
"epoch": 11.46,
"learning_rate": 2.300227575706241e-06,
"loss": 2.6546,
"step": 1100
},
{
"epoch": 11.51,
"learning_rate": 2.2775831730459057e-06,
"loss": 2.6222,
"step": 1105
},
{
"epoch": 11.56,
"learning_rate": 2.2549571491760985e-06,
"loss": 2.5022,
"step": 1110
},
{
"epoch": 11.61,
"learning_rate": 2.232351373734827e-06,
"loss": 2.4277,
"step": 1115
},
{
"epoch": 11.67,
"learning_rate": 2.2097677146869242e-06,
"loss": 2.6073,
"step": 1120
},
{
"epoch": 11.72,
"learning_rate": 2.1872080381697023e-06,
"loss": 2.631,
"step": 1125
},
{
"epoch": 11.77,
"learning_rate": 2.1646742083387397e-06,
"loss": 2.4211,
"step": 1130
},
{
"epoch": 11.82,
"learning_rate": 2.1421680872138483e-06,
"loss": 2.5914,
"step": 1135
},
{
"epoch": 11.88,
"learning_rate": 2.1196915345252085e-06,
"loss": 2.637,
"step": 1140
},
{
"epoch": 11.93,
"learning_rate": 2.0972464075596964e-06,
"loss": 2.6463,
"step": 1145
},
{
"epoch": 11.98,
"learning_rate": 2.0748345610074115e-06,
"loss": 2.6607,
"step": 1150
},
{
"epoch": 12.0,
"eval_loss": 2.597095251083374,
"eval_runtime": 173.3734,
"eval_samples_per_second": 4.418,
"eval_steps_per_second": 1.107,
"step": 1152
},
{
"epoch": 12.03,
"learning_rate": 2.052457846808419e-06,
"loss": 2.6955,
"step": 1155
},
{
"epoch": 12.08,
"learning_rate": 2.0301181139997206e-06,
"loss": 2.578,
"step": 1160
},
{
"epoch": 12.14,
"learning_rate": 2.007817208562466e-06,
"loss": 2.5149,
"step": 1165
},
{
"epoch": 12.19,
"learning_rate": 1.985556973269413e-06,
"loss": 2.6151,
"step": 1170
},
{
"epoch": 12.24,
"learning_rate": 1.9633392475326563e-06,
"loss": 2.6298,
"step": 1175
},
{
"epoch": 12.29,
"learning_rate": 1.9411658672516316e-06,
"loss": 2.6063,
"step": 1180
},
{
"epoch": 12.34,
"learning_rate": 1.919038664661414e-06,
"loss": 2.7178,
"step": 1185
},
{
"epoch": 12.4,
"learning_rate": 1.8969594681813141e-06,
"loss": 2.6825,
"step": 1190
},
{
"epoch": 12.45,
"learning_rate": 1.8749301022637923e-06,
"loss": 2.637,
"step": 1195
},
{
"epoch": 12.5,
"learning_rate": 1.852952387243698e-06,
"loss": 2.5857,
"step": 1200
},
{
"epoch": 12.55,
"learning_rate": 1.8310281391878582e-06,
"loss": 2.5111,
"step": 1205
},
{
"epoch": 12.6,
"learning_rate": 1.8091591697450044e-06,
"loss": 2.5882,
"step": 1210
},
{
"epoch": 12.66,
"learning_rate": 1.7873472859960738e-06,
"loss": 2.5988,
"step": 1215
},
{
"epoch": 12.71,
"learning_rate": 1.765594290304889e-06,
"loss": 2.5168,
"step": 1220
},
{
"epoch": 12.76,
"learning_rate": 1.7439019801692252e-06,
"loss": 2.7292,
"step": 1225
},
{
"epoch": 12.81,
"learning_rate": 1.722272148072273e-06,
"loss": 2.5684,
"step": 1230
},
{
"epoch": 12.86,
"learning_rate": 1.7007065813345286e-06,
"loss": 2.5859,
"step": 1235
},
{
"epoch": 12.92,
"learning_rate": 1.6792070619660977e-06,
"loss": 2.5379,
"step": 1240
},
{
"epoch": 12.97,
"learning_rate": 1.6577753665194502e-06,
"loss": 2.5324,
"step": 1245
},
{
"epoch": 13.0,
"eval_loss": 2.597085475921631,
"eval_runtime": 173.4189,
"eval_samples_per_second": 4.417,
"eval_steps_per_second": 1.107,
"step": 1248
},
{
"epoch": 13.02,
"learning_rate": 1.6364132659426145e-06,
"loss": 2.5232,
"step": 1250
},
{
"epoch": 13.07,
"learning_rate": 1.6151225254328418e-06,
"loss": 2.6904,
"step": 1255
},
{
"epoch": 13.12,
"learning_rate": 1.5939049042907463e-06,
"loss": 2.6439,
"step": 1260
},
{
"epoch": 13.18,
"learning_rate": 1.5727621557749261e-06,
"loss": 2.5261,
"step": 1265
},
{
"epoch": 13.23,
"learning_rate": 1.5516960269570918e-06,
"loss": 2.5726,
"step": 1270
},
{
"epoch": 13.28,
"learning_rate": 1.5307082585776984e-06,
"loss": 2.6521,
"step": 1275
},
{
"epoch": 13.33,
"learning_rate": 1.509800584902108e-06,
"loss": 2.5809,
"step": 1280
},
{
"epoch": 13.39,
"learning_rate": 1.48897473357728e-06,
"loss": 2.5779,
"step": 1285
},
{
"epoch": 13.44,
"learning_rate": 1.4682324254890135e-06,
"loss": 2.5948,
"step": 1290
},
{
"epoch": 13.49,
"learning_rate": 1.4475753746197468e-06,
"loss": 2.5909,
"step": 1295
},
{
"epoch": 13.54,
"learning_rate": 1.427005287906928e-06,
"loss": 2.5414,
"step": 1300
},
{
"epoch": 13.59,
"learning_rate": 1.4065238651019646e-06,
"loss": 2.5634,
"step": 1305
},
{
"epoch": 13.65,
"learning_rate": 1.3861327986297719e-06,
"loss": 2.6414,
"step": 1310
},
{
"epoch": 13.7,
"learning_rate": 1.3658337734489225e-06,
"loss": 2.5816,
"step": 1315
},
{
"epoch": 13.75,
"learning_rate": 1.3456284669124159e-06,
"loss": 2.7111,
"step": 1320
},
{
"epoch": 13.8,
"learning_rate": 1.3255185486290727e-06,
"loss": 2.5295,
"step": 1325
},
{
"epoch": 13.85,
"learning_rate": 1.305505680325575e-06,
"loss": 2.6368,
"step": 1330
},
{
"epoch": 13.91,
"learning_rate": 1.2855915157091498e-06,
"loss": 2.6506,
"step": 1335
},
{
"epoch": 13.96,
"learning_rate": 1.2657777003309244e-06,
"loss": 2.5472,
"step": 1340
},
{
"epoch": 14.0,
"eval_loss": 2.5970234870910645,
"eval_runtime": 173.1837,
"eval_samples_per_second": 4.423,
"eval_steps_per_second": 1.109,
"step": 1344
},
{
"epoch": 14.01,
"learning_rate": 1.2460658714499462e-06,
"loss": 2.5883,
"step": 1345
},
{
"epoch": 14.06,
"learning_rate": 1.2264576578978956e-06,
"loss": 2.5137,
"step": 1350
},
{
"epoch": 14.11,
"learning_rate": 1.2069546799444903e-06,
"loss": 2.5562,
"step": 1355
},
{
"epoch": 14.17,
"learning_rate": 1.1875585491636e-06,
"loss": 2.6397,
"step": 1360
},
{
"epoch": 14.22,
"learning_rate": 1.1682708683000776e-06,
"loss": 2.6961,
"step": 1365
},
{
"epoch": 14.27,
"learning_rate": 1.1490932311373219e-06,
"loss": 2.7012,
"step": 1370
},
{
"epoch": 14.32,
"learning_rate": 1.1300272223655776e-06,
"loss": 2.5606,
"step": 1375
},
{
"epoch": 14.38,
"learning_rate": 1.1110744174509952e-06,
"loss": 2.6933,
"step": 1380
},
{
"epoch": 14.43,
"learning_rate": 1.0922363825054355e-06,
"loss": 2.5044,
"step": 1385
},
{
"epoch": 14.48,
"learning_rate": 1.073514674157068e-06,
"loss": 2.6375,
"step": 1390
},
{
"epoch": 14.53,
"learning_rate": 1.054910839421742e-06,
"loss": 2.5697,
"step": 1395
},
{
"epoch": 14.58,
"learning_rate": 1.0364264155751489e-06,
"loss": 2.6638,
"step": 1400
},
{
"epoch": 14.64,
"learning_rate": 1.018062930025797e-06,
"loss": 2.5092,
"step": 1405
},
{
"epoch": 14.69,
"learning_rate": 9.99821900188798e-07,
"loss": 2.6309,
"step": 1410
},
{
"epoch": 14.74,
"learning_rate": 9.81704833360479e-07,
"loss": 2.6016,
"step": 1415
},
{
"epoch": 14.79,
"learning_rate": 9.637132265938314e-07,
"loss": 2.6111,
"step": 1420
},
{
"epoch": 14.84,
"learning_rate": 9.458485665748071e-07,
"loss": 2.5727,
"step": 1425
},
{
"epoch": 14.9,
"learning_rate": 9.281123294994684e-07,
"loss": 2.5992,
"step": 1430
},
{
"epoch": 14.95,
"learning_rate": 9.105059809520103e-07,
"loss": 2.5187,
"step": 1435
},
{
"epoch": 15.0,
"learning_rate": 8.930309757836517e-07,
"loss": 2.539,
"step": 1440
},
{
"epoch": 15.0,
"eval_loss": 2.597039222717285,
"eval_runtime": 173.4464,
"eval_samples_per_second": 4.416,
"eval_steps_per_second": 1.107,
"step": 1440
},
{
"epoch": 15.05,
"learning_rate": 8.7568875799242e-07,
"loss": 2.5516,
"step": 1445
},
{
"epoch": 15.1,
"learning_rate": 8.584807606038306e-07,
"loss": 2.5809,
"step": 1450
},
{
"epoch": 15.16,
"learning_rate": 8.414084055524727e-07,
"loss": 2.7022,
"step": 1455
},
{
"epoch": 15.21,
"learning_rate": 8.244731035645107e-07,
"loss": 2.5839,
"step": 1460
},
{
"epoch": 15.26,
"learning_rate": 8.076762540411145e-07,
"loss": 2.5779,
"step": 1465
},
{
"epoch": 15.31,
"learning_rate": 7.910192449428216e-07,
"loss": 2.446,
"step": 1470
},
{
"epoch": 15.36,
"learning_rate": 7.74503452674851e-07,
"loss": 2.6257,
"step": 1475
},
{
"epoch": 15.42,
"learning_rate": 7.581302419733633e-07,
"loss": 2.6683,
"step": 1480
},
{
"epoch": 15.47,
"learning_rate": 7.419009657926898e-07,
"loss": 2.6346,
"step": 1485
},
{
"epoch": 15.52,
"learning_rate": 7.258169651935382e-07,
"loss": 2.5915,
"step": 1490
},
{
"epoch": 15.57,
"learning_rate": 7.098795692321775e-07,
"loss": 2.6015,
"step": 1495
},
{
"epoch": 15.62,
"learning_rate": 6.940900948506113e-07,
"loss": 2.5214,
"step": 1500
},
{
"epoch": 15.68,
"learning_rate": 6.784498467677597e-07,
"loss": 2.6015,
"step": 1505
},
{
"epoch": 15.73,
"learning_rate": 6.629601173716454e-07,
"loss": 2.726,
"step": 1510
},
{
"epoch": 15.78,
"learning_rate": 6.476221866126029e-07,
"loss": 2.5489,
"step": 1515
},
{
"epoch": 15.83,
"learning_rate": 6.324373218975105e-07,
"loss": 2.5079,
"step": 1520
},
{
"epoch": 15.89,
"learning_rate": 6.174067779850645e-07,
"loss": 2.6254,
"step": 1525
},
{
"epoch": 15.94,
"learning_rate": 6.025317968820954e-07,
"loss": 2.6369,
"step": 1530
},
{
"epoch": 15.99,
"learning_rate": 5.878136077409357e-07,
"loss": 2.5757,
"step": 1535
},
{
"epoch": 16.0,
"eval_loss": 2.597059965133667,
"eval_runtime": 173.4779,
"eval_samples_per_second": 4.416,
"eval_steps_per_second": 1.107,
"step": 1536
},
{
"epoch": 16.04,
"learning_rate": 5.732534267578549e-07,
"loss": 2.6266,
"step": 1540
},
{
"epoch": 16.09,
"learning_rate": 5.588524570725612e-07,
"loss": 2.6665,
"step": 1545
},
{
"epoch": 16.15,
"learning_rate": 5.44611888668784e-07,
"loss": 2.5927,
"step": 1550
},
{
"epoch": 16.2,
"learning_rate": 5.305328982759431e-07,
"loss": 2.6236,
"step": 1555
},
{
"epoch": 16.25,
"learning_rate": 5.166166492719124e-07,
"loss": 2.4952,
"step": 1560
},
{
"epoch": 16.3,
"learning_rate": 5.028642915868869e-07,
"loss": 2.6098,
"step": 1565
},
{
"epoch": 16.35,
"learning_rate": 4.892769616083648e-07,
"loss": 2.6309,
"step": 1570
},
{
"epoch": 16.41,
"learning_rate": 4.7585578208723976e-07,
"loss": 2.4235,
"step": 1575
},
{
"epoch": 16.46,
"learning_rate": 4.626018620450309e-07,
"loss": 2.6291,
"step": 1580
},
{
"epoch": 16.51,
"learning_rate": 4.4951629668223583e-07,
"loss": 2.6039,
"step": 1585
},
{
"epoch": 16.56,
"learning_rate": 4.366001672878406e-07,
"loss": 2.5953,
"step": 1590
},
{
"epoch": 16.61,
"learning_rate": 4.23854541149962e-07,
"loss": 2.539,
"step": 1595
},
{
"epoch": 16.67,
"learning_rate": 4.1128047146765936e-07,
"loss": 2.7039,
"step": 1600
},
{
"epoch": 16.72,
"learning_rate": 3.9887899726390455e-07,
"loss": 2.7045,
"step": 1605
},
{
"epoch": 16.77,
"learning_rate": 3.8665114329972997e-07,
"loss": 2.4848,
"step": 1610
},
{
"epoch": 16.82,
"learning_rate": 3.7459791998954137e-07,
"loss": 2.6097,
"step": 1615
},
{
"epoch": 16.88,
"learning_rate": 3.627203233176341e-07,
"loss": 2.5642,
"step": 1620
},
{
"epoch": 16.93,
"learning_rate": 3.510193347558866e-07,
"loss": 2.6055,
"step": 1625
},
{
"epoch": 16.98,
"learning_rate": 3.3949592118266426e-07,
"loss": 2.6495,
"step": 1630
},
{
"epoch": 17.0,
"eval_loss": 2.5970218181610107,
"eval_runtime": 173.4401,
"eval_samples_per_second": 4.417,
"eval_steps_per_second": 1.107,
"step": 1632
},
{
"epoch": 17.03,
"learning_rate": 3.2815103480291934e-07,
"loss": 2.6929,
"step": 1635
},
{
"epoch": 17.08,
"learning_rate": 3.1698561306951065e-07,
"loss": 2.5553,
"step": 1640
},
{
"epoch": 17.14,
"learning_rate": 3.0600057860574015e-07,
"loss": 2.6909,
"step": 1645
},
{
"epoch": 17.19,
"learning_rate": 2.9519683912911267e-07,
"loss": 2.5462,
"step": 1650
},
{
"epoch": 17.24,
"learning_rate": 2.8457528737633163e-07,
"loss": 2.6499,
"step": 1655
},
{
"epoch": 17.29,
"learning_rate": 2.7413680102952755e-07,
"loss": 2.5658,
"step": 1660
},
{
"epoch": 17.34,
"learning_rate": 2.638822426437371e-07,
"loss": 2.6463,
"step": 1665
},
{
"epoch": 17.4,
"learning_rate": 2.5381245957562304e-07,
"loss": 2.5734,
"step": 1670
},
{
"epoch": 17.45,
"learning_rate": 2.439282839134602e-07,
"loss": 2.4343,
"step": 1675
},
{
"epoch": 17.5,
"learning_rate": 2.3423053240837518e-07,
"loss": 2.6411,
"step": 1680
},
{
"epoch": 17.55,
"learning_rate": 2.247200064068572e-07,
"loss": 2.626,
"step": 1685
},
{
"epoch": 17.6,
"learning_rate": 2.1539749178454223e-07,
"loss": 2.6194,
"step": 1690
},
{
"epoch": 17.66,
"learning_rate": 2.0626375888127187e-07,
"loss": 2.672,
"step": 1695
},
{
"epoch": 17.71,
"learning_rate": 1.9731956243744106e-07,
"loss": 2.6633,
"step": 1700
},
{
"epoch": 17.76,
"learning_rate": 1.8856564153163143e-07,
"loss": 2.5198,
"step": 1705
},
{
"epoch": 17.81,
"learning_rate": 1.800027195195389e-07,
"loss": 2.5005,
"step": 1710
},
{
"epoch": 17.86,
"learning_rate": 1.7163150397420087e-07,
"loss": 2.5845,
"step": 1715
},
{
"epoch": 17.92,
"learning_rate": 1.6345268662752904e-07,
"loss": 2.6347,
"step": 1720
},
{
"epoch": 17.97,
"learning_rate": 1.5546694331315086e-07,
"loss": 2.5647,
"step": 1725
},
{
"epoch": 18.0,
"eval_loss": 2.597024917602539,
"eval_runtime": 173.3767,
"eval_samples_per_second": 4.418,
"eval_steps_per_second": 1.107,
"step": 1728
},
{
"epoch": 18.02,
"learning_rate": 1.4767493391056153e-07,
"loss": 2.639,
"step": 1730
},
{
"epoch": 18.07,
"learning_rate": 1.400773022905985e-07,
"loss": 2.5828,
"step": 1735
},
{
"epoch": 18.12,
"learning_rate": 1.3267467626223606e-07,
"loss": 2.6745,
"step": 1740
},
{
"epoch": 18.18,
"learning_rate": 1.2546766752070893e-07,
"loss": 2.6199,
"step": 1745
},
{
"epoch": 18.23,
"learning_rate": 1.1845687159696584e-07,
"loss": 2.5725,
"step": 1750
},
{
"epoch": 18.28,
"learning_rate": 1.1164286780845995e-07,
"loss": 2.6234,
"step": 1755
},
{
"epoch": 18.33,
"learning_rate": 1.0502621921127776e-07,
"loss": 2.6208,
"step": 1760
},
{
"epoch": 18.39,
"learning_rate": 9.860747255361485e-08,
"loss": 2.5672,
"step": 1765
},
{
"epoch": 18.44,
"learning_rate": 9.238715823059324e-08,
"loss": 2.6916,
"step": 1770
},
{
"epoch": 18.49,
"learning_rate": 8.636579024043717e-08,
"loss": 2.6221,
"step": 1775
},
{
"epoch": 18.54,
"learning_rate": 8.054386614199772e-08,
"loss": 2.5351,
"step": 1780
},
{
"epoch": 18.59,
"learning_rate": 7.492186701364007e-08,
"loss": 2.5847,
"step": 1785
},
{
"epoch": 18.65,
"learning_rate": 6.950025741349037e-08,
"loss": 2.5973,
"step": 1790
},
{
"epoch": 18.7,
"learning_rate": 6.427948534104777e-08,
"loss": 2.6242,
"step": 1795
},
{
"epoch": 18.75,
"learning_rate": 5.92599822001666e-08,
"loss": 2.5065,
"step": 1800
},
{
"epoch": 18.8,
"learning_rate": 5.444216276340702e-08,
"loss": 2.6884,
"step": 1805
},
{
"epoch": 18.85,
"learning_rate": 4.982642513776226e-08,
"loss": 2.5535,
"step": 1810
},
{
"epoch": 18.91,
"learning_rate": 4.541315073176178e-08,
"loss": 2.4669,
"step": 1815
},
{
"epoch": 18.96,
"learning_rate": 4.120270422395572e-08,
"loss": 2.5605,
"step": 1820
},
{
"epoch": 19.0,
"eval_loss": 2.5970375537872314,
"eval_runtime": 173.322,
"eval_samples_per_second": 4.42,
"eval_steps_per_second": 1.108,
"step": 1824
},
{
"epoch": 19.01,
"learning_rate": 3.7195433532778713e-08,
"loss": 2.6154,
"step": 1825
},
{
"epoch": 19.06,
"learning_rate": 3.339166978780256e-08,
"loss": 2.6572,
"step": 1830
},
{
"epoch": 19.11,
"learning_rate": 2.9791727302373175e-08,
"loss": 2.5629,
"step": 1835
},
{
"epoch": 19.17,
"learning_rate": 2.6395903547638825e-08,
"loss": 2.6195,
"step": 1840
},
{
"epoch": 19.22,
"learning_rate": 2.3204479127968415e-08,
"loss": 2.6312,
"step": 1845
},
{
"epoch": 19.27,
"learning_rate": 2.0217717757766152e-08,
"loss": 2.6655,
"step": 1850
},
{
"epoch": 19.32,
"learning_rate": 1.7435866239678135e-08,
"loss": 2.5804,
"step": 1855
},
{
"epoch": 19.38,
"learning_rate": 1.4859154444200885e-08,
"loss": 2.6275,
"step": 1860
},
{
"epoch": 19.43,
"learning_rate": 1.2487795290684889e-08,
"loss": 2.6881,
"step": 1865
},
{
"epoch": 19.48,
"learning_rate": 1.0321984729741163e-08,
"loss": 2.5813,
"step": 1870
},
{
"epoch": 19.53,
"learning_rate": 8.361901727049204e-09,
"loss": 2.6291,
"step": 1875
},
{
"epoch": 19.58,
"learning_rate": 6.607708248569378e-09,
"loss": 2.5822,
"step": 1880
},
{
"epoch": 19.64,
"learning_rate": 5.0595492471583465e-09,
"loss": 2.4705,
"step": 1885
},
{
"epoch": 19.69,
"learning_rate": 3.71755265059226e-09,
"loss": 2.5638,
"step": 1890
},
{
"epoch": 19.74,
"learning_rate": 2.581829350994658e-09,
"loss": 2.4875,
"step": 1895
},
{
"epoch": 19.79,
"learning_rate": 1.6524731956749085e-09,
"loss": 2.607,
"step": 1900
},
{
"epoch": 19.84,
"learning_rate": 9.295609793708027e-10,
"loss": 2.4936,
"step": 1905
},
{
"epoch": 19.9,
"learning_rate": 4.13152437906128e-10,
"loss": 2.6588,
"step": 1910
},
{
"epoch": 19.95,
"learning_rate": 1.0329024325128611e-10,
"loss": 2.5804,
"step": 1915
},
{
"epoch": 20.0,
"learning_rate": 0.0,
"loss": 2.6608,
"step": 1920
},
{
"epoch": 20.0,
"eval_loss": 2.5970427989959717,
"eval_runtime": 173.461,
"eval_samples_per_second": 4.416,
"eval_steps_per_second": 1.107,
"step": 1920
},
{
"epoch": 20.0,
"step": 1920,
"total_flos": 1.0984887148766822e+18,
"train_loss": 2.604488531500101,
"train_runtime": 14508.1949,
"train_samples_per_second": 1.056,
"train_steps_per_second": 0.132
}
],
"logging_steps": 5,
"max_steps": 1920,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 50,
"total_flos": 1.0984887148766822e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}