coding_llamaduo_60k_v0.2 / trainer_state.json
chansung's picture
Model save
989ee20 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 10.0,
"eval_steps": 500,
"global_step": 1260,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.007936507936507936,
"grad_norm": 300.0,
"learning_rate": 1.5873015873015873e-06,
"loss": 44.0514,
"step": 1
},
{
"epoch": 0.03968253968253968,
"grad_norm": 294.0,
"learning_rate": 7.936507936507936e-06,
"loss": 44.4732,
"step": 5
},
{
"epoch": 0.07936507936507936,
"grad_norm": 172.0,
"learning_rate": 1.5873015873015872e-05,
"loss": 39.0478,
"step": 10
},
{
"epoch": 0.11904761904761904,
"grad_norm": 61.5,
"learning_rate": 2.380952380952381e-05,
"loss": 27.5033,
"step": 15
},
{
"epoch": 0.15873015873015872,
"grad_norm": 19.625,
"learning_rate": 3.1746031746031745e-05,
"loss": 19.8771,
"step": 20
},
{
"epoch": 0.1984126984126984,
"grad_norm": 12.6875,
"learning_rate": 3.968253968253968e-05,
"loss": 18.3177,
"step": 25
},
{
"epoch": 0.23809523809523808,
"grad_norm": 5.875,
"learning_rate": 4.761904761904762e-05,
"loss": 16.5284,
"step": 30
},
{
"epoch": 0.2777777777777778,
"grad_norm": 3.015625,
"learning_rate": 5.555555555555556e-05,
"loss": 15.6508,
"step": 35
},
{
"epoch": 0.31746031746031744,
"grad_norm": 3.828125,
"learning_rate": 6.349206349206349e-05,
"loss": 14.7333,
"step": 40
},
{
"epoch": 0.35714285714285715,
"grad_norm": 7.34375,
"learning_rate": 7.142857142857143e-05,
"loss": 14.189,
"step": 45
},
{
"epoch": 0.3968253968253968,
"grad_norm": 19.0,
"learning_rate": 7.936507936507937e-05,
"loss": 11.8511,
"step": 50
},
{
"epoch": 0.4365079365079365,
"grad_norm": 16.625,
"learning_rate": 8.730158730158731e-05,
"loss": 5.6454,
"step": 55
},
{
"epoch": 0.47619047619047616,
"grad_norm": 2.203125,
"learning_rate": 9.523809523809524e-05,
"loss": 1.7925,
"step": 60
},
{
"epoch": 0.5158730158730159,
"grad_norm": 2.265625,
"learning_rate": 0.00010317460317460319,
"loss": 1.4147,
"step": 65
},
{
"epoch": 0.5555555555555556,
"grad_norm": 0.74609375,
"learning_rate": 0.00011111111111111112,
"loss": 1.2306,
"step": 70
},
{
"epoch": 0.5952380952380952,
"grad_norm": 2.046875,
"learning_rate": 0.00011904761904761905,
"loss": 1.1237,
"step": 75
},
{
"epoch": 0.6349206349206349,
"grad_norm": 2.0,
"learning_rate": 0.00012698412698412698,
"loss": 1.0467,
"step": 80
},
{
"epoch": 0.6746031746031746,
"grad_norm": 7.90625,
"learning_rate": 0.00013492063492063494,
"loss": 0.9706,
"step": 85
},
{
"epoch": 0.7142857142857143,
"grad_norm": 1.5078125,
"learning_rate": 0.00014285714285714287,
"loss": 0.944,
"step": 90
},
{
"epoch": 0.753968253968254,
"grad_norm": 2.765625,
"learning_rate": 0.0001507936507936508,
"loss": 0.9101,
"step": 95
},
{
"epoch": 0.7936507936507936,
"grad_norm": 2.140625,
"learning_rate": 0.00015873015873015873,
"loss": 0.8892,
"step": 100
},
{
"epoch": 0.8333333333333334,
"grad_norm": 1.640625,
"learning_rate": 0.0001666666666666667,
"loss": 0.8475,
"step": 105
},
{
"epoch": 0.873015873015873,
"grad_norm": 27.125,
"learning_rate": 0.00017460317460317462,
"loss": 0.8386,
"step": 110
},
{
"epoch": 0.9126984126984127,
"grad_norm": 1.8828125,
"learning_rate": 0.00018253968253968255,
"loss": 0.8187,
"step": 115
},
{
"epoch": 0.9523809523809523,
"grad_norm": 1.6015625,
"learning_rate": 0.00019047619047619048,
"loss": 0.7724,
"step": 120
},
{
"epoch": 0.9920634920634921,
"grad_norm": 1.6640625,
"learning_rate": 0.00019841269841269844,
"loss": 0.7499,
"step": 125
},
{
"epoch": 1.0,
"eval_loss": 1.2579786777496338,
"eval_runtime": 1.0146,
"eval_samples_per_second": 1.971,
"eval_steps_per_second": 0.986,
"step": 126
},
{
"epoch": 1.0317460317460316,
"grad_norm": 0.453125,
"learning_rate": 0.00019999386012995552,
"loss": 0.7176,
"step": 130
},
{
"epoch": 1.0714285714285714,
"grad_norm": 0.5078125,
"learning_rate": 0.00019996891820008164,
"loss": 0.7074,
"step": 135
},
{
"epoch": 1.1111111111111112,
"grad_norm": 0.609375,
"learning_rate": 0.00019992479525042303,
"loss": 0.6984,
"step": 140
},
{
"epoch": 1.1507936507936507,
"grad_norm": 7.375,
"learning_rate": 0.0001998614997468427,
"loss": 0.6891,
"step": 145
},
{
"epoch": 1.1904761904761905,
"grad_norm": 0.90625,
"learning_rate": 0.0001997790438338385,
"loss": 0.6864,
"step": 150
},
{
"epoch": 1.2301587301587302,
"grad_norm": 0.306640625,
"learning_rate": 0.00019967744333221278,
"loss": 0.6766,
"step": 155
},
{
"epoch": 1.2698412698412698,
"grad_norm": 8.1875,
"learning_rate": 0.00019955671773603696,
"loss": 0.6776,
"step": 160
},
{
"epoch": 1.3095238095238095,
"grad_norm": 2.203125,
"learning_rate": 0.0001994168902089112,
"loss": 0.6565,
"step": 165
},
{
"epoch": 1.3492063492063493,
"grad_norm": 0.80859375,
"learning_rate": 0.00019925798757952,
"loss": 0.6495,
"step": 170
},
{
"epoch": 1.3888888888888888,
"grad_norm": 1.90625,
"learning_rate": 0.00019908004033648453,
"loss": 0.6625,
"step": 175
},
{
"epoch": 1.4285714285714286,
"grad_norm": 1.0234375,
"learning_rate": 0.00019888308262251285,
"loss": 0.6444,
"step": 180
},
{
"epoch": 1.4682539682539684,
"grad_norm": 0.9609375,
"learning_rate": 0.00019866715222784895,
"loss": 0.635,
"step": 185
},
{
"epoch": 1.507936507936508,
"grad_norm": 1.953125,
"learning_rate": 0.0001984322905830219,
"loss": 0.6417,
"step": 190
},
{
"epoch": 1.5476190476190477,
"grad_norm": 1.0859375,
"learning_rate": 0.0001981785427508966,
"loss": 0.6381,
"step": 195
},
{
"epoch": 1.5873015873015874,
"grad_norm": 0.69140625,
"learning_rate": 0.00019790595741802757,
"loss": 0.6256,
"step": 200
},
{
"epoch": 1.626984126984127,
"grad_norm": 0.30078125,
"learning_rate": 0.00019761458688531756,
"loss": 0.6247,
"step": 205
},
{
"epoch": 1.6666666666666665,
"grad_norm": 0.921875,
"learning_rate": 0.00019730448705798239,
"loss": 0.6244,
"step": 210
},
{
"epoch": 1.7063492063492065,
"grad_norm": 0.6328125,
"learning_rate": 0.0001969757174348246,
"loss": 0.6094,
"step": 215
},
{
"epoch": 1.746031746031746,
"grad_norm": 0.76953125,
"learning_rate": 0.0001966283410968174,
"loss": 0.6156,
"step": 220
},
{
"epoch": 1.7857142857142856,
"grad_norm": 1.1484375,
"learning_rate": 0.0001962624246950012,
"loss": 0.6037,
"step": 225
},
{
"epoch": 1.8253968253968254,
"grad_norm": 0.984375,
"learning_rate": 0.0001958780384376955,
"loss": 0.6068,
"step": 230
},
{
"epoch": 1.8650793650793651,
"grad_norm": 0.640625,
"learning_rate": 0.00019547525607702774,
"loss": 0.5994,
"step": 235
},
{
"epoch": 1.9047619047619047,
"grad_norm": 1.8203125,
"learning_rate": 0.0001950541548947829,
"loss": 0.6115,
"step": 240
},
{
"epoch": 1.9444444444444444,
"grad_norm": 0.30078125,
"learning_rate": 0.00019461481568757506,
"loss": 0.598,
"step": 245
},
{
"epoch": 1.9841269841269842,
"grad_norm": 0.396484375,
"learning_rate": 0.00019415732275134513,
"loss": 0.6058,
"step": 250
},
{
"epoch": 2.0,
"eval_loss": 1.1687145233154297,
"eval_runtime": 1.0157,
"eval_samples_per_second": 1.969,
"eval_steps_per_second": 0.985,
"step": 252
},
{
"epoch": 2.0238095238095237,
"grad_norm": 0.458984375,
"learning_rate": 0.0001936817638651871,
"loss": 0.5677,
"step": 255
},
{
"epoch": 2.0634920634920633,
"grad_norm": 0.5078125,
"learning_rate": 0.0001931882302745057,
"loss": 0.5648,
"step": 260
},
{
"epoch": 2.1031746031746033,
"grad_norm": 0.66796875,
"learning_rate": 0.00019267681667350928,
"loss": 0.5502,
"step": 265
},
{
"epoch": 2.142857142857143,
"grad_norm": 0.48828125,
"learning_rate": 0.00019214762118704076,
"loss": 0.5573,
"step": 270
},
{
"epoch": 2.1825396825396823,
"grad_norm": 0.26953125,
"learning_rate": 0.00019160074535175058,
"loss": 0.5622,
"step": 275
},
{
"epoch": 2.2222222222222223,
"grad_norm": 0.291015625,
"learning_rate": 0.0001910362940966147,
"loss": 0.5586,
"step": 280
},
{
"epoch": 2.261904761904762,
"grad_norm": 0.3984375,
"learning_rate": 0.00019045437572280194,
"loss": 0.5545,
"step": 285
},
{
"epoch": 2.3015873015873014,
"grad_norm": 0.34765625,
"learning_rate": 0.0001898551018828944,
"loss": 0.5489,
"step": 290
},
{
"epoch": 2.3412698412698414,
"grad_norm": 0.275390625,
"learning_rate": 0.0001892385875594645,
"loss": 0.5577,
"step": 295
},
{
"epoch": 2.380952380952381,
"grad_norm": 0.2578125,
"learning_rate": 0.00018860495104301345,
"loss": 0.5462,
"step": 300
},
{
"epoch": 2.4206349206349205,
"grad_norm": 0.38671875,
"learning_rate": 0.0001879543139092747,
"loss": 0.557,
"step": 305
},
{
"epoch": 2.4603174603174605,
"grad_norm": 0.44921875,
"learning_rate": 0.00018728680099588748,
"loss": 0.5531,
"step": 310
},
{
"epoch": 2.5,
"grad_norm": 0.27734375,
"learning_rate": 0.00018660254037844388,
"loss": 0.5597,
"step": 315
},
{
"epoch": 2.5396825396825395,
"grad_norm": 0.71484375,
"learning_rate": 0.00018590166334591531,
"loss": 0.5578,
"step": 320
},
{
"epoch": 2.5793650793650795,
"grad_norm": 0.375,
"learning_rate": 0.000185184304375462,
"loss": 0.5546,
"step": 325
},
{
"epoch": 2.619047619047619,
"grad_norm": 0.2109375,
"learning_rate": 0.0001844506011066308,
"loss": 0.5532,
"step": 330
},
{
"epoch": 2.6587301587301586,
"grad_norm": 0.349609375,
"learning_rate": 0.00018370069431494646,
"loss": 0.5509,
"step": 335
},
{
"epoch": 2.6984126984126986,
"grad_norm": 0.27734375,
"learning_rate": 0.00018293472788490095,
"loss": 0.5479,
"step": 340
},
{
"epoch": 2.738095238095238,
"grad_norm": 0.279296875,
"learning_rate": 0.00018215284878234642,
"loss": 0.5505,
"step": 345
},
{
"epoch": 2.7777777777777777,
"grad_norm": 0.28125,
"learning_rate": 0.00018135520702629675,
"loss": 0.5466,
"step": 350
},
{
"epoch": 2.817460317460317,
"grad_norm": 0.373046875,
"learning_rate": 0.0001805419556601437,
"loss": 0.5548,
"step": 355
},
{
"epoch": 2.857142857142857,
"grad_norm": 0.25,
"learning_rate": 0.00017971325072229226,
"loss": 0.5521,
"step": 360
},
{
"epoch": 2.8968253968253967,
"grad_norm": 0.318359375,
"learning_rate": 0.0001788692512162216,
"loss": 0.5413,
"step": 365
},
{
"epoch": 2.9365079365079367,
"grad_norm": 0.390625,
"learning_rate": 0.00017801011907997725,
"loss": 0.5546,
"step": 370
},
{
"epoch": 2.9761904761904763,
"grad_norm": 0.26171875,
"learning_rate": 0.0001771360191551,
"loss": 0.5571,
"step": 375
},
{
"epoch": 3.0,
"eval_loss": 1.1491789817810059,
"eval_runtime": 1.0142,
"eval_samples_per_second": 1.972,
"eval_steps_per_second": 0.986,
"step": 378
},
{
"epoch": 3.015873015873016,
"grad_norm": 0.30859375,
"learning_rate": 0.00017624711915499764,
"loss": 0.5262,
"step": 380
},
{
"epoch": 3.0555555555555554,
"grad_norm": 0.2734375,
"learning_rate": 0.00017534358963276607,
"loss": 0.5035,
"step": 385
},
{
"epoch": 3.0952380952380953,
"grad_norm": 0.203125,
"learning_rate": 0.00017442560394846516,
"loss": 0.5017,
"step": 390
},
{
"epoch": 3.134920634920635,
"grad_norm": 0.21484375,
"learning_rate": 0.00017349333823585617,
"loss": 0.5052,
"step": 395
},
{
"epoch": 3.1746031746031744,
"grad_norm": 0.232421875,
"learning_rate": 0.00017254697136860703,
"loss": 0.5056,
"step": 400
},
{
"epoch": 3.2142857142857144,
"grad_norm": 0.333984375,
"learning_rate": 0.00017158668492597186,
"loss": 0.5199,
"step": 405
},
{
"epoch": 3.253968253968254,
"grad_norm": 0.2216796875,
"learning_rate": 0.00017061266315795146,
"loss": 0.5038,
"step": 410
},
{
"epoch": 3.2936507936507935,
"grad_norm": 0.333984375,
"learning_rate": 0.0001696250929499412,
"loss": 0.501,
"step": 415
},
{
"epoch": 3.3333333333333335,
"grad_norm": 0.79296875,
"learning_rate": 0.0001686241637868734,
"loss": 0.5044,
"step": 420
},
{
"epoch": 3.373015873015873,
"grad_norm": 0.4296875,
"learning_rate": 0.0001676100677168608,
"loss": 0.4998,
"step": 425
},
{
"epoch": 3.4126984126984126,
"grad_norm": 0.265625,
"learning_rate": 0.00016658299931434858,
"loss": 0.5172,
"step": 430
},
{
"epoch": 3.4523809523809526,
"grad_norm": 0.212890625,
"learning_rate": 0.000165543155642781,
"loss": 0.5082,
"step": 435
},
{
"epoch": 3.492063492063492,
"grad_norm": 0.22265625,
"learning_rate": 0.00016449073621679127,
"loss": 0.5016,
"step": 440
},
{
"epoch": 3.5317460317460316,
"grad_norm": 0.1865234375,
"learning_rate": 0.0001634259429639203,
"loss": 0.5109,
"step": 445
},
{
"epoch": 3.571428571428571,
"grad_norm": 0.197265625,
"learning_rate": 0.00016234898018587337,
"loss": 0.5178,
"step": 450
},
{
"epoch": 3.611111111111111,
"grad_norm": 0.267578125,
"learning_rate": 0.0001612600545193203,
"loss": 0.5163,
"step": 455
},
{
"epoch": 3.6507936507936507,
"grad_norm": 0.29296875,
"learning_rate": 0.00016015937489624848,
"loss": 0.5078,
"step": 460
},
{
"epoch": 3.6904761904761907,
"grad_norm": 0.25,
"learning_rate": 0.00015904715250387498,
"loss": 0.508,
"step": 465
},
{
"epoch": 3.7301587301587302,
"grad_norm": 0.2890625,
"learning_rate": 0.00015792360074412613,
"loss": 0.5055,
"step": 470
},
{
"epoch": 3.7698412698412698,
"grad_norm": 0.19140625,
"learning_rate": 0.00015678893519269197,
"loss": 0.5083,
"step": 475
},
{
"epoch": 3.8095238095238093,
"grad_norm": 0.322265625,
"learning_rate": 0.00015564337355766412,
"loss": 0.5121,
"step": 480
},
{
"epoch": 3.8492063492063493,
"grad_norm": 0.2109375,
"learning_rate": 0.00015448713563776374,
"loss": 0.4984,
"step": 485
},
{
"epoch": 3.888888888888889,
"grad_norm": 0.412109375,
"learning_rate": 0.00015332044328016914,
"loss": 0.5044,
"step": 490
},
{
"epoch": 3.928571428571429,
"grad_norm": 0.44140625,
"learning_rate": 0.0001521435203379498,
"loss": 0.5127,
"step": 495
},
{
"epoch": 3.9682539682539684,
"grad_norm": 0.54296875,
"learning_rate": 0.0001509565926271159,
"loss": 0.5118,
"step": 500
},
{
"epoch": 4.0,
"eval_loss": 1.155090570449829,
"eval_runtime": 1.0165,
"eval_samples_per_second": 1.968,
"eval_steps_per_second": 0.984,
"step": 504
},
{
"epoch": 4.007936507936508,
"grad_norm": 0.97265625,
"learning_rate": 0.00014975988788329064,
"loss": 0.4977,
"step": 505
},
{
"epoch": 4.0476190476190474,
"grad_norm": 0.515625,
"learning_rate": 0.00014855363571801523,
"loss": 0.4642,
"step": 510
},
{
"epoch": 4.087301587301587,
"grad_norm": 0.251953125,
"learning_rate": 0.00014733806757469286,
"loss": 0.457,
"step": 515
},
{
"epoch": 4.1269841269841265,
"grad_norm": 0.51953125,
"learning_rate": 0.000146113416684182,
"loss": 0.4674,
"step": 520
},
{
"epoch": 4.166666666666667,
"grad_norm": 0.259765625,
"learning_rate": 0.00014487991802004623,
"loss": 0.4608,
"step": 525
},
{
"epoch": 4.2063492063492065,
"grad_norm": 0.45703125,
"learning_rate": 0.00014363780825347005,
"loss": 0.4601,
"step": 530
},
{
"epoch": 4.246031746031746,
"grad_norm": 0.408203125,
"learning_rate": 0.00014238732570784866,
"loss": 0.4656,
"step": 535
},
{
"epoch": 4.285714285714286,
"grad_norm": 0.263671875,
"learning_rate": 0.00014112871031306119,
"loss": 0.4661,
"step": 540
},
{
"epoch": 4.325396825396825,
"grad_norm": 0.42578125,
"learning_rate": 0.00013986220355943494,
"loss": 0.4652,
"step": 545
},
{
"epoch": 4.365079365079365,
"grad_norm": 0.5,
"learning_rate": 0.00013858804845141116,
"loss": 0.4667,
"step": 550
},
{
"epoch": 4.404761904761905,
"grad_norm": 0.205078125,
"learning_rate": 0.0001373064894609194,
"loss": 0.469,
"step": 555
},
{
"epoch": 4.444444444444445,
"grad_norm": 0.59375,
"learning_rate": 0.00013601777248047105,
"loss": 0.4654,
"step": 560
},
{
"epoch": 4.484126984126984,
"grad_norm": 0.298828125,
"learning_rate": 0.00013472214477597977,
"loss": 0.4662,
"step": 565
},
{
"epoch": 4.523809523809524,
"grad_norm": 0.28515625,
"learning_rate": 0.00013341985493931877,
"loss": 0.4669,
"step": 570
},
{
"epoch": 4.563492063492063,
"grad_norm": 0.3125,
"learning_rate": 0.00013211115284062335,
"loss": 0.465,
"step": 575
},
{
"epoch": 4.603174603174603,
"grad_norm": 0.30859375,
"learning_rate": 0.00013079628958034855,
"loss": 0.4696,
"step": 580
},
{
"epoch": 4.642857142857143,
"grad_norm": 0.38671875,
"learning_rate": 0.00012947551744109043,
"loss": 0.4837,
"step": 585
},
{
"epoch": 4.682539682539683,
"grad_norm": 0.75390625,
"learning_rate": 0.00012814908983918073,
"loss": 0.4752,
"step": 590
},
{
"epoch": 4.722222222222222,
"grad_norm": 0.59375,
"learning_rate": 0.00012681726127606376,
"loss": 0.4678,
"step": 595
},
{
"epoch": 4.761904761904762,
"grad_norm": 0.306640625,
"learning_rate": 0.0001254802872894655,
"loss": 0.4753,
"step": 600
},
{
"epoch": 4.801587301587301,
"grad_norm": 0.333984375,
"learning_rate": 0.00012413842440436333,
"loss": 0.473,
"step": 605
},
{
"epoch": 4.841269841269841,
"grad_norm": 0.353515625,
"learning_rate": 0.000122791930083767,
"loss": 0.4646,
"step": 610
},
{
"epoch": 4.880952380952381,
"grad_norm": 0.224609375,
"learning_rate": 0.00012144106267931876,
"loss": 0.4715,
"step": 615
},
{
"epoch": 4.920634920634921,
"grad_norm": 0.2734375,
"learning_rate": 0.00012008608138172393,
"loss": 0.4704,
"step": 620
},
{
"epoch": 4.9603174603174605,
"grad_norm": 0.31640625,
"learning_rate": 0.00011872724617101969,
"loss": 0.4657,
"step": 625
},
{
"epoch": 5.0,
"grad_norm": 0.224609375,
"learning_rate": 0.00011736481776669306,
"loss": 0.4711,
"step": 630
},
{
"epoch": 5.0,
"eval_loss": 1.1766771078109741,
"eval_runtime": 1.015,
"eval_samples_per_second": 1.971,
"eval_steps_per_second": 0.985,
"step": 630
},
{
"epoch": 5.0396825396825395,
"grad_norm": 0.2890625,
"learning_rate": 0.0001159990575776563,
"loss": 0.4228,
"step": 635
},
{
"epoch": 5.079365079365079,
"grad_norm": 0.3125,
"learning_rate": 0.00011463022765209088,
"loss": 0.4163,
"step": 640
},
{
"epoch": 5.119047619047619,
"grad_norm": 0.353515625,
"learning_rate": 0.00011325859062716795,
"loss": 0.4213,
"step": 645
},
{
"epoch": 5.158730158730159,
"grad_norm": 0.365234375,
"learning_rate": 0.00011188440967865641,
"loss": 0.4319,
"step": 650
},
{
"epoch": 5.198412698412699,
"grad_norm": 0.3203125,
"learning_rate": 0.00011050794847042731,
"loss": 0.4258,
"step": 655
},
{
"epoch": 5.238095238095238,
"grad_norm": 0.26953125,
"learning_rate": 0.00010912947110386484,
"loss": 0.4237,
"step": 660
},
{
"epoch": 5.277777777777778,
"grad_norm": 0.5078125,
"learning_rate": 0.0001077492420671931,
"loss": 0.4238,
"step": 665
},
{
"epoch": 5.317460317460317,
"grad_norm": 0.22265625,
"learning_rate": 0.00010636752618472887,
"loss": 0.4225,
"step": 670
},
{
"epoch": 5.357142857142857,
"grad_norm": 0.2451171875,
"learning_rate": 0.00010498458856606972,
"loss": 0.4214,
"step": 675
},
{
"epoch": 5.396825396825397,
"grad_norm": 0.283203125,
"learning_rate": 0.00010360069455522765,
"loss": 0.425,
"step": 680
},
{
"epoch": 5.436507936507937,
"grad_norm": 0.43359375,
"learning_rate": 0.00010221610967971735,
"loss": 0.4281,
"step": 685
},
{
"epoch": 5.476190476190476,
"grad_norm": 0.275390625,
"learning_rate": 0.00010083109959960973,
"loss": 0.427,
"step": 690
},
{
"epoch": 5.515873015873016,
"grad_norm": 0.2099609375,
"learning_rate": 9.944593005655947e-05,
"loss": 0.4299,
"step": 695
},
{
"epoch": 5.555555555555555,
"grad_norm": 0.216796875,
"learning_rate": 9.806086682281758e-05,
"loss": 0.4413,
"step": 700
},
{
"epoch": 5.595238095238095,
"grad_norm": 0.2158203125,
"learning_rate": 9.667617565023735e-05,
"loss": 0.4352,
"step": 705
},
{
"epoch": 5.634920634920634,
"grad_norm": 0.2578125,
"learning_rate": 9.529212221928483e-05,
"loss": 0.4337,
"step": 710
},
{
"epoch": 5.674603174603175,
"grad_norm": 0.2294921875,
"learning_rate": 9.390897208806266e-05,
"loss": 0.4242,
"step": 715
},
{
"epoch": 5.714285714285714,
"grad_norm": 0.365234375,
"learning_rate": 9.252699064135758e-05,
"loss": 0.4286,
"step": 720
},
{
"epoch": 5.753968253968254,
"grad_norm": 0.5078125,
"learning_rate": 9.114644303972096e-05,
"loss": 0.4349,
"step": 725
},
{
"epoch": 5.7936507936507935,
"grad_norm": 0.318359375,
"learning_rate": 8.976759416859256e-05,
"loss": 0.4311,
"step": 730
},
{
"epoch": 5.833333333333333,
"grad_norm": 0.2470703125,
"learning_rate": 8.839070858747697e-05,
"loss": 0.4257,
"step": 735
},
{
"epoch": 5.8730158730158735,
"grad_norm": 0.236328125,
"learning_rate": 8.701605047918276e-05,
"loss": 0.4332,
"step": 740
},
{
"epoch": 5.912698412698413,
"grad_norm": 0.2255859375,
"learning_rate": 8.564388359913356e-05,
"loss": 0.4309,
"step": 745
},
{
"epoch": 5.9523809523809526,
"grad_norm": 0.2138671875,
"learning_rate": 8.427447122476148e-05,
"loss": 0.4311,
"step": 750
},
{
"epoch": 5.992063492063492,
"grad_norm": 0.244140625,
"learning_rate": 8.290807610499206e-05,
"loss": 0.4287,
"step": 755
},
{
"epoch": 6.0,
"eval_loss": 1.1948192119598389,
"eval_runtime": 1.0151,
"eval_samples_per_second": 1.97,
"eval_steps_per_second": 0.985,
"step": 756
},
{
"epoch": 6.031746031746032,
"grad_norm": 0.265625,
"learning_rate": 8.154496040983073e-05,
"loss": 0.3917,
"step": 760
},
{
"epoch": 6.071428571428571,
"grad_norm": 0.322265625,
"learning_rate": 8.018538568006027e-05,
"loss": 0.3851,
"step": 765
},
{
"epoch": 6.111111111111111,
"grad_norm": 0.267578125,
"learning_rate": 7.882961277705895e-05,
"loss": 0.3846,
"step": 770
},
{
"epoch": 6.150793650793651,
"grad_norm": 0.392578125,
"learning_rate": 7.747790183274922e-05,
"loss": 0.3897,
"step": 775
},
{
"epoch": 6.190476190476191,
"grad_norm": 0.25,
"learning_rate": 7.613051219968623e-05,
"loss": 0.3894,
"step": 780
},
{
"epoch": 6.23015873015873,
"grad_norm": 0.30859375,
"learning_rate": 7.478770240129579e-05,
"loss": 0.386,
"step": 785
},
{
"epoch": 6.26984126984127,
"grad_norm": 0.29296875,
"learning_rate": 7.344973008227161e-05,
"loss": 0.383,
"step": 790
},
{
"epoch": 6.309523809523809,
"grad_norm": 0.265625,
"learning_rate": 7.211685195914097e-05,
"loss": 0.3867,
"step": 795
},
{
"epoch": 6.349206349206349,
"grad_norm": 0.263671875,
"learning_rate": 7.078932377100877e-05,
"loss": 0.393,
"step": 800
},
{
"epoch": 6.388888888888889,
"grad_norm": 0.345703125,
"learning_rate": 6.94674002304887e-05,
"loss": 0.3856,
"step": 805
},
{
"epoch": 6.428571428571429,
"grad_norm": 0.2255859375,
"learning_rate": 6.815133497483157e-05,
"loss": 0.397,
"step": 810
},
{
"epoch": 6.468253968253968,
"grad_norm": 0.302734375,
"learning_rate": 6.684138051726012e-05,
"loss": 0.3879,
"step": 815
},
{
"epoch": 6.507936507936508,
"grad_norm": 0.248046875,
"learning_rate": 6.553778819851926e-05,
"loss": 0.3852,
"step": 820
},
{
"epoch": 6.5476190476190474,
"grad_norm": 0.24609375,
"learning_rate": 6.424080813865138e-05,
"loss": 0.3956,
"step": 825
},
{
"epoch": 6.587301587301587,
"grad_norm": 0.4296875,
"learning_rate": 6.295068918900586e-05,
"loss": 0.394,
"step": 830
},
{
"epoch": 6.6269841269841265,
"grad_norm": 0.2392578125,
"learning_rate": 6.16676788844919e-05,
"loss": 0.3904,
"step": 835
},
{
"epoch": 6.666666666666667,
"grad_norm": 0.33984375,
"learning_rate": 6.039202339608432e-05,
"loss": 0.395,
"step": 840
},
{
"epoch": 6.7063492063492065,
"grad_norm": 0.326171875,
"learning_rate": 5.912396748359046e-05,
"loss": 0.3892,
"step": 845
},
{
"epoch": 6.746031746031746,
"grad_norm": 0.337890625,
"learning_rate": 5.786375444868828e-05,
"loss": 0.3945,
"step": 850
},
{
"epoch": 6.785714285714286,
"grad_norm": 0.443359375,
"learning_rate": 5.6611626088244194e-05,
"loss": 0.3948,
"step": 855
},
{
"epoch": 6.825396825396825,
"grad_norm": 0.314453125,
"learning_rate": 5.5367822647919424e-05,
"loss": 0.3953,
"step": 860
},
{
"epoch": 6.865079365079366,
"grad_norm": 0.283203125,
"learning_rate": 5.4132582776074126e-05,
"loss": 0.3983,
"step": 865
},
{
"epoch": 6.904761904761905,
"grad_norm": 0.236328125,
"learning_rate": 5.290614347797802e-05,
"loss": 0.3862,
"step": 870
},
{
"epoch": 6.944444444444445,
"grad_norm": 0.2431640625,
"learning_rate": 5.168874007033615e-05,
"loss": 0.39,
"step": 875
},
{
"epoch": 6.984126984126984,
"grad_norm": 0.232421875,
"learning_rate": 5.048060613613888e-05,
"loss": 0.3943,
"step": 880
},
{
"epoch": 7.0,
"eval_loss": 1.2383077144622803,
"eval_runtime": 1.0141,
"eval_samples_per_second": 1.972,
"eval_steps_per_second": 0.986,
"step": 882
},
{
"epoch": 7.023809523809524,
"grad_norm": 0.2216796875,
"learning_rate": 4.92819734798441e-05,
"loss": 0.3718,
"step": 885
},
{
"epoch": 7.063492063492063,
"grad_norm": 0.41015625,
"learning_rate": 4.809307208290114e-05,
"loss": 0.3505,
"step": 890
},
{
"epoch": 7.103174603174603,
"grad_norm": 0.236328125,
"learning_rate": 4.691413005962415e-05,
"loss": 0.3559,
"step": 895
},
{
"epoch": 7.142857142857143,
"grad_norm": 0.248046875,
"learning_rate": 4.574537361342407e-05,
"loss": 0.3581,
"step": 900
},
{
"epoch": 7.182539682539683,
"grad_norm": 0.224609375,
"learning_rate": 4.458702699340667e-05,
"loss": 0.3601,
"step": 905
},
{
"epoch": 7.222222222222222,
"grad_norm": 0.259765625,
"learning_rate": 4.343931245134616e-05,
"loss": 0.3587,
"step": 910
},
{
"epoch": 7.261904761904762,
"grad_norm": 0.240234375,
"learning_rate": 4.23024501990417e-05,
"loss": 0.3589,
"step": 915
},
{
"epoch": 7.301587301587301,
"grad_norm": 0.26171875,
"learning_rate": 4.117665836606549e-05,
"loss": 0.3595,
"step": 920
},
{
"epoch": 7.341269841269841,
"grad_norm": 0.259765625,
"learning_rate": 4.00621529579101e-05,
"loss": 0.359,
"step": 925
},
{
"epoch": 7.380952380952381,
"grad_norm": 0.24609375,
"learning_rate": 3.89591478145437e-05,
"loss": 0.3637,
"step": 930
},
{
"epoch": 7.420634920634921,
"grad_norm": 0.26953125,
"learning_rate": 3.786785456938049e-05,
"loss": 0.3667,
"step": 935
},
{
"epoch": 7.4603174603174605,
"grad_norm": 0.26171875,
"learning_rate": 3.6788482608674826e-05,
"loss": 0.3566,
"step": 940
},
{
"epoch": 7.5,
"grad_norm": 0.248046875,
"learning_rate": 3.5721239031346066e-05,
"loss": 0.3538,
"step": 945
},
{
"epoch": 7.5396825396825395,
"grad_norm": 0.2392578125,
"learning_rate": 3.4666328609242725e-05,
"loss": 0.3564,
"step": 950
},
{
"epoch": 7.579365079365079,
"grad_norm": 0.2421875,
"learning_rate": 3.362395374785283e-05,
"loss": 0.3588,
"step": 955
},
{
"epoch": 7.619047619047619,
"grad_norm": 0.25390625,
"learning_rate": 3.259431444746846e-05,
"loss": 0.3617,
"step": 960
},
{
"epoch": 7.658730158730159,
"grad_norm": 0.2470703125,
"learning_rate": 3.157760826481174e-05,
"loss": 0.3616,
"step": 965
},
{
"epoch": 7.698412698412699,
"grad_norm": 0.25,
"learning_rate": 3.057403027512963e-05,
"loss": 0.3531,
"step": 970
},
{
"epoch": 7.738095238095238,
"grad_norm": 0.287109375,
"learning_rate": 2.9583773034764826e-05,
"loss": 0.3547,
"step": 975
},
{
"epoch": 7.777777777777778,
"grad_norm": 0.2421875,
"learning_rate": 2.8607026544210114e-05,
"loss": 0.3609,
"step": 980
},
{
"epoch": 7.817460317460317,
"grad_norm": 0.2392578125,
"learning_rate": 2.764397821165292e-05,
"loss": 0.3588,
"step": 985
},
{
"epoch": 7.857142857142857,
"grad_norm": 0.271484375,
"learning_rate": 2.669481281701739e-05,
"loss": 0.3648,
"step": 990
},
{
"epoch": 7.896825396825397,
"grad_norm": 0.251953125,
"learning_rate": 2.5759712476510622e-05,
"loss": 0.3635,
"step": 995
},
{
"epoch": 7.936507936507937,
"grad_norm": 0.2412109375,
"learning_rate": 2.4838856607680183e-05,
"loss": 0.3568,
"step": 1000
},
{
"epoch": 7.976190476190476,
"grad_norm": 0.25390625,
"learning_rate": 2.3932421894989167e-05,
"loss": 0.3612,
"step": 1005
},
{
"epoch": 8.0,
"eval_loss": 1.2904332876205444,
"eval_runtime": 1.0153,
"eval_samples_per_second": 1.97,
"eval_steps_per_second": 0.985,
"step": 1008
},
{
"epoch": 8.015873015873016,
"grad_norm": 0.26171875,
"learning_rate": 2.304058225591581e-05,
"loss": 0.354,
"step": 1010
},
{
"epoch": 8.055555555555555,
"grad_norm": 0.23828125,
"learning_rate": 2.2163508807583998e-05,
"loss": 0.3464,
"step": 1015
},
{
"epoch": 8.095238095238095,
"grad_norm": 0.259765625,
"learning_rate": 2.1301369833931117e-05,
"loss": 0.3411,
"step": 1020
},
{
"epoch": 8.134920634920634,
"grad_norm": 0.259765625,
"learning_rate": 2.045433075341927e-05,
"loss": 0.3369,
"step": 1025
},
{
"epoch": 8.174603174603174,
"grad_norm": 0.232421875,
"learning_rate": 1.962255408729662e-05,
"loss": 0.3369,
"step": 1030
},
{
"epoch": 8.214285714285714,
"grad_norm": 0.2353515625,
"learning_rate": 1.880619942841435e-05,
"loss": 0.3391,
"step": 1035
},
{
"epoch": 8.253968253968253,
"grad_norm": 0.2470703125,
"learning_rate": 1.8005423410605772e-05,
"loss": 0.3395,
"step": 1040
},
{
"epoch": 8.293650793650794,
"grad_norm": 0.244140625,
"learning_rate": 1.7220379678632814e-05,
"loss": 0.3409,
"step": 1045
},
{
"epoch": 8.333333333333334,
"grad_norm": 0.2412109375,
"learning_rate": 1.6451218858706374e-05,
"loss": 0.3483,
"step": 1050
},
{
"epoch": 8.373015873015873,
"grad_norm": 0.2421875,
"learning_rate": 1.5698088529585597e-05,
"loss": 0.3459,
"step": 1055
},
{
"epoch": 8.412698412698413,
"grad_norm": 0.2412109375,
"learning_rate": 1.49611331942621e-05,
"loss": 0.336,
"step": 1060
},
{
"epoch": 8.452380952380953,
"grad_norm": 0.2578125,
"learning_rate": 1.4240494252234049e-05,
"loss": 0.3349,
"step": 1065
},
{
"epoch": 8.492063492063492,
"grad_norm": 0.2412109375,
"learning_rate": 1.3536309972375948e-05,
"loss": 0.3463,
"step": 1070
},
{
"epoch": 8.531746031746032,
"grad_norm": 0.23828125,
"learning_rate": 1.2848715466408967e-05,
"loss": 0.3372,
"step": 1075
},
{
"epoch": 8.571428571428571,
"grad_norm": 0.2314453125,
"learning_rate": 1.2177842662977135e-05,
"loss": 0.3346,
"step": 1080
},
{
"epoch": 8.61111111111111,
"grad_norm": 0.236328125,
"learning_rate": 1.1523820282334219e-05,
"loss": 0.3449,
"step": 1085
},
{
"epoch": 8.65079365079365,
"grad_norm": 0.234375,
"learning_rate": 1.088677381164609e-05,
"loss": 0.3368,
"step": 1090
},
{
"epoch": 8.69047619047619,
"grad_norm": 0.23828125,
"learning_rate": 1.0266825480913611e-05,
"loss": 0.3379,
"step": 1095
},
{
"epoch": 8.73015873015873,
"grad_norm": 0.2421875,
"learning_rate": 9.664094239520372e-06,
"loss": 0.348,
"step": 1100
},
{
"epoch": 8.76984126984127,
"grad_norm": 0.244140625,
"learning_rate": 9.07869573340987e-06,
"loss": 0.343,
"step": 1105
},
{
"epoch": 8.80952380952381,
"grad_norm": 0.236328125,
"learning_rate": 8.510742282896544e-06,
"loss": 0.3396,
"step": 1110
},
{
"epoch": 8.84920634920635,
"grad_norm": 0.23828125,
"learning_rate": 7.960342861114921e-06,
"loss": 0.3391,
"step": 1115
},
{
"epoch": 8.88888888888889,
"grad_norm": 0.2333984375,
"learning_rate": 7.427603073110967e-06,
"loss": 0.3405,
"step": 1120
},
{
"epoch": 8.928571428571429,
"grad_norm": 0.2314453125,
"learning_rate": 6.9126251355795864e-06,
"loss": 0.3378,
"step": 1125
},
{
"epoch": 8.968253968253968,
"grad_norm": 0.263671875,
"learning_rate": 6.415507857252389e-06,
"loss": 0.3457,
"step": 1130
},
{
"epoch": 9.0,
"eval_loss": 1.3253473043441772,
"eval_runtime": 1.0138,
"eval_samples_per_second": 1.973,
"eval_steps_per_second": 0.986,
"step": 1134
},
{
"epoch": 9.007936507936508,
"grad_norm": 0.232421875,
"learning_rate": 5.936346619939271e-06,
"loss": 0.3402,
"step": 1135
},
{
"epoch": 9.047619047619047,
"grad_norm": 0.248046875,
"learning_rate": 5.475233360227516e-06,
"loss": 0.3362,
"step": 1140
},
{
"epoch": 9.087301587301587,
"grad_norm": 0.232421875,
"learning_rate": 5.03225655184194e-06,
"loss": 0.3346,
"step": 1145
},
{
"epoch": 9.126984126984127,
"grad_norm": 0.228515625,
"learning_rate": 4.607501188669394e-06,
"loss": 0.3358,
"step": 1150
},
{
"epoch": 9.166666666666666,
"grad_norm": 0.232421875,
"learning_rate": 4.20104876845111e-06,
"loss": 0.333,
"step": 1155
},
{
"epoch": 9.206349206349206,
"grad_norm": 0.2373046875,
"learning_rate": 3.8129772771456797e-06,
"loss": 0.3393,
"step": 1160
},
{
"epoch": 9.246031746031745,
"grad_norm": 0.232421875,
"learning_rate": 3.4433611739658645e-06,
"loss": 0.3409,
"step": 1165
},
{
"epoch": 9.285714285714286,
"grad_norm": 0.2373046875,
"learning_rate": 3.092271377092215e-06,
"loss": 0.34,
"step": 1170
},
{
"epoch": 9.325396825396826,
"grad_norm": 0.2353515625,
"learning_rate": 2.759775250065899e-06,
"loss": 0.3362,
"step": 1175
},
{
"epoch": 9.365079365079366,
"grad_norm": 0.23828125,
"learning_rate": 2.4459365888638062e-06,
"loss": 0.333,
"step": 1180
},
{
"epoch": 9.404761904761905,
"grad_norm": 0.2314453125,
"learning_rate": 2.150815609657875e-06,
"loss": 0.3315,
"step": 1185
},
{
"epoch": 9.444444444444445,
"grad_norm": 0.23046875,
"learning_rate": 1.874468937261531e-06,
"loss": 0.3371,
"step": 1190
},
{
"epoch": 9.484126984126984,
"grad_norm": 0.24609375,
"learning_rate": 1.6169495942650714e-06,
"loss": 0.339,
"step": 1195
},
{
"epoch": 9.523809523809524,
"grad_norm": 0.23046875,
"learning_rate": 1.378306990862177e-06,
"loss": 0.3413,
"step": 1200
},
{
"epoch": 9.563492063492063,
"grad_norm": 0.2333984375,
"learning_rate": 1.158586915369675e-06,
"loss": 0.3313,
"step": 1205
},
{
"epoch": 9.603174603174603,
"grad_norm": 0.2333984375,
"learning_rate": 9.578315254420767e-07,
"loss": 0.339,
"step": 1210
},
{
"epoch": 9.642857142857142,
"grad_norm": 0.2392578125,
"learning_rate": 7.760793399827937e-07,
"loss": 0.3284,
"step": 1215
},
{
"epoch": 9.682539682539682,
"grad_norm": 0.23828125,
"learning_rate": 6.13365231753571e-07,
"loss": 0.3359,
"step": 1220
},
{
"epoch": 9.722222222222221,
"grad_norm": 0.228515625,
"learning_rate": 4.6972042068341714e-07,
"loss": 0.3394,
"step": 1225
},
{
"epoch": 9.761904761904763,
"grad_norm": 0.2373046875,
"learning_rate": 3.451724678784518e-07,
"loss": 0.337,
"step": 1230
},
{
"epoch": 9.801587301587302,
"grad_norm": 0.2421875,
"learning_rate": 2.397452703337577e-07,
"loss": 0.3344,
"step": 1235
},
{
"epoch": 9.841269841269842,
"grad_norm": 0.234375,
"learning_rate": 1.5345905634827074e-07,
"loss": 0.3307,
"step": 1240
},
{
"epoch": 9.880952380952381,
"grad_norm": 0.2373046875,
"learning_rate": 8.633038164358454e-08,
"loss": 0.3361,
"step": 1245
},
{
"epoch": 9.920634920634921,
"grad_norm": 0.2392578125,
"learning_rate": 3.8372126187413704e-08,
"loss": 0.34,
"step": 1250
},
{
"epoch": 9.96031746031746,
"grad_norm": 0.2314453125,
"learning_rate": 9.593491722270642e-09,
"loss": 0.3385,
"step": 1255
},
{
"epoch": 10.0,
"grad_norm": 0.2470703125,
"learning_rate": 0.0,
"loss": 0.3328,
"step": 1260
},
{
"epoch": 10.0,
"eval_loss": 1.332617163658142,
"eval_runtime": 1.0141,
"eval_samples_per_second": 1.972,
"eval_steps_per_second": 0.986,
"step": 1260
},
{
"epoch": 10.0,
"step": 1260,
"total_flos": 3.8899222565240177e+18,
"train_loss": 1.363766341266178,
"train_runtime": 8098.2535,
"train_samples_per_second": 4.97,
"train_steps_per_second": 0.156
}
],
"logging_steps": 5,
"max_steps": 1260,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 100,
"total_flos": 3.8899222565240177e+18,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}