ALM-AHME's picture
End of training
63db5e6
{
"best_metric": 0.9907502569373073,
"best_model_checkpoint": "beit-large-patch16-224-finetuned-Lesion-Classification-HAM10000-AH-60-20-20/checkpoint-1464",
"epoch": 12.0,
"global_step": 1464,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 1.8968133535660093e-08,
"loss": 2.2184,
"step": 5
},
{
"epoch": 0.08,
"learning_rate": 3.7936267071320186e-08,
"loss": 2.166,
"step": 10
},
{
"epoch": 0.12,
"learning_rate": 5.6904400606980276e-08,
"loss": 2.1412,
"step": 15
},
{
"epoch": 0.16,
"learning_rate": 7.587253414264037e-08,
"loss": 2.1034,
"step": 20
},
{
"epoch": 0.2,
"learning_rate": 9.484066767830046e-08,
"loss": 2.1319,
"step": 25
},
{
"epoch": 0.25,
"learning_rate": 1.1380880121396055e-07,
"loss": 2.2418,
"step": 30
},
{
"epoch": 0.29,
"learning_rate": 1.3277693474962064e-07,
"loss": 2.085,
"step": 35
},
{
"epoch": 0.33,
"learning_rate": 1.5174506828528074e-07,
"loss": 2.0375,
"step": 40
},
{
"epoch": 0.37,
"learning_rate": 1.7071320182094083e-07,
"loss": 2.0931,
"step": 45
},
{
"epoch": 0.41,
"learning_rate": 1.896813353566009e-07,
"loss": 2.0902,
"step": 50
},
{
"epoch": 0.45,
"learning_rate": 2.0864946889226102e-07,
"loss": 2.0333,
"step": 55
},
{
"epoch": 0.49,
"learning_rate": 2.276176024279211e-07,
"loss": 1.9706,
"step": 60
},
{
"epoch": 0.53,
"learning_rate": 2.465857359635812e-07,
"loss": 2.0811,
"step": 65
},
{
"epoch": 0.57,
"learning_rate": 2.6555386949924127e-07,
"loss": 2.0207,
"step": 70
},
{
"epoch": 0.61,
"learning_rate": 2.845220030349014e-07,
"loss": 1.9804,
"step": 75
},
{
"epoch": 0.66,
"learning_rate": 3.034901365705615e-07,
"loss": 2.031,
"step": 80
},
{
"epoch": 0.7,
"learning_rate": 3.2245827010622155e-07,
"loss": 2.1196,
"step": 85
},
{
"epoch": 0.74,
"learning_rate": 3.4142640364188166e-07,
"loss": 1.9947,
"step": 90
},
{
"epoch": 0.78,
"learning_rate": 3.6039453717754177e-07,
"loss": 1.9709,
"step": 95
},
{
"epoch": 0.82,
"learning_rate": 3.793626707132018e-07,
"loss": 1.9723,
"step": 100
},
{
"epoch": 0.86,
"learning_rate": 3.9833080424886193e-07,
"loss": 1.9318,
"step": 105
},
{
"epoch": 0.9,
"learning_rate": 4.1729893778452204e-07,
"loss": 1.9495,
"step": 110
},
{
"epoch": 0.94,
"learning_rate": 4.3626707132018215e-07,
"loss": 1.9893,
"step": 115
},
{
"epoch": 0.98,
"learning_rate": 4.552352048558422e-07,
"loss": 1.9688,
"step": 120
},
{
"epoch": 1.0,
"eval_accuracy": 0.2774922918807811,
"eval_loss": 1.8424760103225708,
"eval_runtime": 469.1069,
"eval_samples_per_second": 2.074,
"eval_steps_per_second": 0.13,
"step": 122
},
{
"epoch": 1.02,
"learning_rate": 4.742033383915023e-07,
"loss": 1.9411,
"step": 125
},
{
"epoch": 1.07,
"learning_rate": 4.931714719271624e-07,
"loss": 1.8792,
"step": 130
},
{
"epoch": 1.11,
"learning_rate": 5.121396054628225e-07,
"loss": 1.8279,
"step": 135
},
{
"epoch": 1.15,
"learning_rate": 5.311077389984825e-07,
"loss": 1.8292,
"step": 140
},
{
"epoch": 1.19,
"learning_rate": 5.500758725341426e-07,
"loss": 1.8529,
"step": 145
},
{
"epoch": 1.23,
"learning_rate": 5.690440060698028e-07,
"loss": 1.83,
"step": 150
},
{
"epoch": 1.27,
"learning_rate": 5.880121396054629e-07,
"loss": 1.7656,
"step": 155
},
{
"epoch": 1.31,
"learning_rate": 6.06980273141123e-07,
"loss": 1.718,
"step": 160
},
{
"epoch": 1.35,
"learning_rate": 6.259484066767831e-07,
"loss": 1.7614,
"step": 165
},
{
"epoch": 1.39,
"learning_rate": 6.449165402124431e-07,
"loss": 1.6862,
"step": 170
},
{
"epoch": 1.43,
"learning_rate": 6.638846737481032e-07,
"loss": 1.7277,
"step": 175
},
{
"epoch": 1.48,
"learning_rate": 6.828528072837633e-07,
"loss": 1.7737,
"step": 180
},
{
"epoch": 1.52,
"learning_rate": 7.018209408194235e-07,
"loss": 1.6894,
"step": 185
},
{
"epoch": 1.56,
"learning_rate": 7.207890743550835e-07,
"loss": 1.6661,
"step": 190
},
{
"epoch": 1.6,
"learning_rate": 7.397572078907436e-07,
"loss": 1.7083,
"step": 195
},
{
"epoch": 1.64,
"learning_rate": 7.587253414264036e-07,
"loss": 1.6996,
"step": 200
},
{
"epoch": 1.68,
"learning_rate": 7.776934749620638e-07,
"loss": 1.6233,
"step": 205
},
{
"epoch": 1.72,
"learning_rate": 7.966616084977239e-07,
"loss": 1.6174,
"step": 210
},
{
"epoch": 1.76,
"learning_rate": 8.156297420333841e-07,
"loss": 1.5842,
"step": 215
},
{
"epoch": 1.8,
"learning_rate": 8.345978755690441e-07,
"loss": 1.5328,
"step": 220
},
{
"epoch": 1.84,
"learning_rate": 8.535660091047042e-07,
"loss": 1.5476,
"step": 225
},
{
"epoch": 1.89,
"learning_rate": 8.725341426403643e-07,
"loss": 1.4752,
"step": 230
},
{
"epoch": 1.93,
"learning_rate": 8.915022761760243e-07,
"loss": 1.4618,
"step": 235
},
{
"epoch": 1.97,
"learning_rate": 9.104704097116844e-07,
"loss": 1.4822,
"step": 240
},
{
"epoch": 2.0,
"eval_accuracy": 0.5457348406988695,
"eval_loss": 1.3833471536636353,
"eval_runtime": 38.3426,
"eval_samples_per_second": 25.376,
"eval_steps_per_second": 1.591,
"step": 244
},
{
"epoch": 2.01,
"learning_rate": 9.294385432473445e-07,
"loss": 1.3992,
"step": 245
},
{
"epoch": 2.05,
"learning_rate": 9.484066767830046e-07,
"loss": 1.3256,
"step": 250
},
{
"epoch": 2.09,
"learning_rate": 9.673748103186647e-07,
"loss": 1.377,
"step": 255
},
{
"epoch": 2.13,
"learning_rate": 9.863429438543249e-07,
"loss": 1.4332,
"step": 260
},
{
"epoch": 2.17,
"learning_rate": 1.005311077389985e-06,
"loss": 1.3325,
"step": 265
},
{
"epoch": 2.21,
"learning_rate": 1.024279210925645e-06,
"loss": 1.3756,
"step": 270
},
{
"epoch": 2.25,
"learning_rate": 1.043247344461305e-06,
"loss": 1.2157,
"step": 275
},
{
"epoch": 2.3,
"learning_rate": 1.062215477996965e-06,
"loss": 1.3218,
"step": 280
},
{
"epoch": 2.34,
"learning_rate": 1.0811836115326254e-06,
"loss": 1.2928,
"step": 285
},
{
"epoch": 2.38,
"learning_rate": 1.1001517450682853e-06,
"loss": 1.2055,
"step": 290
},
{
"epoch": 2.42,
"learning_rate": 1.1191198786039454e-06,
"loss": 1.189,
"step": 295
},
{
"epoch": 2.46,
"learning_rate": 1.1380880121396055e-06,
"loss": 1.1767,
"step": 300
},
{
"epoch": 2.5,
"learning_rate": 1.1570561456752656e-06,
"loss": 1.1959,
"step": 305
},
{
"epoch": 2.54,
"learning_rate": 1.1760242792109257e-06,
"loss": 1.1909,
"step": 310
},
{
"epoch": 2.58,
"learning_rate": 1.1949924127465858e-06,
"loss": 1.1548,
"step": 315
},
{
"epoch": 2.62,
"learning_rate": 1.213960546282246e-06,
"loss": 1.2163,
"step": 320
},
{
"epoch": 2.66,
"learning_rate": 1.232928679817906e-06,
"loss": 1.1614,
"step": 325
},
{
"epoch": 2.7,
"learning_rate": 1.2518968133535662e-06,
"loss": 1.1428,
"step": 330
},
{
"epoch": 2.75,
"learning_rate": 1.2708649468892263e-06,
"loss": 1.0076,
"step": 335
},
{
"epoch": 2.79,
"learning_rate": 1.2898330804248862e-06,
"loss": 1.0842,
"step": 340
},
{
"epoch": 2.83,
"learning_rate": 1.3088012139605463e-06,
"loss": 1.0182,
"step": 345
},
{
"epoch": 2.87,
"learning_rate": 1.3277693474962064e-06,
"loss": 1.0125,
"step": 350
},
{
"epoch": 2.91,
"learning_rate": 1.3467374810318665e-06,
"loss": 1.1462,
"step": 355
},
{
"epoch": 2.95,
"learning_rate": 1.3657056145675266e-06,
"loss": 1.0148,
"step": 360
},
{
"epoch": 2.99,
"learning_rate": 1.384673748103187e-06,
"loss": 1.1239,
"step": 365
},
{
"epoch": 3.0,
"eval_accuracy": 0.6680369989722508,
"eval_loss": 0.9320738315582275,
"eval_runtime": 38.3484,
"eval_samples_per_second": 25.373,
"eval_steps_per_second": 1.591,
"step": 366
},
{
"epoch": 3.03,
"learning_rate": 1.403641881638847e-06,
"loss": 0.9881,
"step": 370
},
{
"epoch": 3.07,
"learning_rate": 1.422610015174507e-06,
"loss": 1.0889,
"step": 375
},
{
"epoch": 3.11,
"learning_rate": 1.441578148710167e-06,
"loss": 0.9645,
"step": 380
},
{
"epoch": 3.16,
"learning_rate": 1.4605462822458272e-06,
"loss": 0.9425,
"step": 385
},
{
"epoch": 3.2,
"learning_rate": 1.4795144157814873e-06,
"loss": 0.8992,
"step": 390
},
{
"epoch": 3.24,
"learning_rate": 1.4984825493171474e-06,
"loss": 0.9565,
"step": 395
},
{
"epoch": 3.28,
"learning_rate": 1.5174506828528073e-06,
"loss": 0.8908,
"step": 400
},
{
"epoch": 3.32,
"learning_rate": 1.5364188163884674e-06,
"loss": 0.885,
"step": 405
},
{
"epoch": 3.36,
"learning_rate": 1.5553869499241275e-06,
"loss": 0.9257,
"step": 410
},
{
"epoch": 3.4,
"learning_rate": 1.5743550834597876e-06,
"loss": 0.8028,
"step": 415
},
{
"epoch": 3.44,
"learning_rate": 1.5933232169954477e-06,
"loss": 0.8498,
"step": 420
},
{
"epoch": 3.48,
"learning_rate": 1.6122913505311078e-06,
"loss": 0.8806,
"step": 425
},
{
"epoch": 3.52,
"learning_rate": 1.6312594840667682e-06,
"loss": 0.9644,
"step": 430
},
{
"epoch": 3.57,
"learning_rate": 1.6502276176024283e-06,
"loss": 0.9284,
"step": 435
},
{
"epoch": 3.61,
"learning_rate": 1.6691957511380882e-06,
"loss": 0.7972,
"step": 440
},
{
"epoch": 3.65,
"learning_rate": 1.6881638846737483e-06,
"loss": 0.7791,
"step": 445
},
{
"epoch": 3.69,
"learning_rate": 1.7071320182094084e-06,
"loss": 0.7934,
"step": 450
},
{
"epoch": 3.73,
"learning_rate": 1.7261001517450685e-06,
"loss": 0.7144,
"step": 455
},
{
"epoch": 3.77,
"learning_rate": 1.7450682852807286e-06,
"loss": 0.8308,
"step": 460
},
{
"epoch": 3.81,
"learning_rate": 1.7640364188163885e-06,
"loss": 0.8119,
"step": 465
},
{
"epoch": 3.85,
"learning_rate": 1.7830045523520486e-06,
"loss": 0.7736,
"step": 470
},
{
"epoch": 3.89,
"learning_rate": 1.8019726858877087e-06,
"loss": 0.7132,
"step": 475
},
{
"epoch": 3.93,
"learning_rate": 1.8209408194233688e-06,
"loss": 0.7348,
"step": 480
},
{
"epoch": 3.98,
"learning_rate": 1.839908952959029e-06,
"loss": 0.8686,
"step": 485
},
{
"epoch": 4.0,
"eval_accuracy": 0.7697841726618705,
"eval_loss": 0.6690807938575745,
"eval_runtime": 38.3878,
"eval_samples_per_second": 25.347,
"eval_steps_per_second": 1.589,
"step": 488
},
{
"epoch": 4.02,
"learning_rate": 1.858877086494689e-06,
"loss": 0.7654,
"step": 490
},
{
"epoch": 4.06,
"learning_rate": 1.8778452200303494e-06,
"loss": 0.7116,
"step": 495
},
{
"epoch": 4.1,
"learning_rate": 1.8968133535660093e-06,
"loss": 0.731,
"step": 500
},
{
"epoch": 4.14,
"learning_rate": 1.9157814871016696e-06,
"loss": 0.6989,
"step": 505
},
{
"epoch": 4.18,
"learning_rate": 1.9347496206373295e-06,
"loss": 0.7768,
"step": 510
},
{
"epoch": 4.22,
"learning_rate": 1.9537177541729894e-06,
"loss": 0.7992,
"step": 515
},
{
"epoch": 4.26,
"learning_rate": 1.9726858877086497e-06,
"loss": 0.7129,
"step": 520
},
{
"epoch": 4.3,
"learning_rate": 1.9916540212443096e-06,
"loss": 0.7175,
"step": 525
},
{
"epoch": 4.34,
"learning_rate": 2.01062215477997e-06,
"loss": 0.6502,
"step": 530
},
{
"epoch": 4.39,
"learning_rate": 2.02959028831563e-06,
"loss": 0.7056,
"step": 535
},
{
"epoch": 4.43,
"learning_rate": 2.04855842185129e-06,
"loss": 0.6236,
"step": 540
},
{
"epoch": 4.47,
"learning_rate": 2.06752655538695e-06,
"loss": 0.6581,
"step": 545
},
{
"epoch": 4.51,
"learning_rate": 2.08649468892261e-06,
"loss": 0.6582,
"step": 550
},
{
"epoch": 4.55,
"learning_rate": 2.1054628224582703e-06,
"loss": 0.6542,
"step": 555
},
{
"epoch": 4.59,
"learning_rate": 2.12443095599393e-06,
"loss": 0.6086,
"step": 560
},
{
"epoch": 4.63,
"learning_rate": 2.1433990895295905e-06,
"loss": 0.5802,
"step": 565
},
{
"epoch": 4.67,
"learning_rate": 2.162367223065251e-06,
"loss": 0.6414,
"step": 570
},
{
"epoch": 4.71,
"learning_rate": 2.1813353566009107e-06,
"loss": 0.6911,
"step": 575
},
{
"epoch": 4.75,
"learning_rate": 2.2003034901365706e-06,
"loss": 0.6478,
"step": 580
},
{
"epoch": 4.8,
"learning_rate": 2.219271623672231e-06,
"loss": 0.5478,
"step": 585
},
{
"epoch": 4.84,
"learning_rate": 2.238239757207891e-06,
"loss": 0.5731,
"step": 590
},
{
"epoch": 4.88,
"learning_rate": 2.257207890743551e-06,
"loss": 0.5935,
"step": 595
},
{
"epoch": 4.92,
"learning_rate": 2.276176024279211e-06,
"loss": 0.5285,
"step": 600
},
{
"epoch": 4.96,
"learning_rate": 2.2951441578148714e-06,
"loss": 0.6257,
"step": 605
},
{
"epoch": 5.0,
"learning_rate": 2.3141122913505313e-06,
"loss": 0.5234,
"step": 610
},
{
"epoch": 5.0,
"eval_accuracy": 0.8335046248715313,
"eval_loss": 0.48716917634010315,
"eval_runtime": 38.3444,
"eval_samples_per_second": 25.375,
"eval_steps_per_second": 1.591,
"step": 610
},
{
"epoch": 5.04,
"learning_rate": 2.333080424886191e-06,
"loss": 0.5729,
"step": 615
},
{
"epoch": 5.08,
"learning_rate": 2.3520485584218515e-06,
"loss": 0.5729,
"step": 620
},
{
"epoch": 5.12,
"learning_rate": 2.3710166919575114e-06,
"loss": 0.5388,
"step": 625
},
{
"epoch": 5.16,
"learning_rate": 2.3899848254931717e-06,
"loss": 0.5941,
"step": 630
},
{
"epoch": 5.2,
"learning_rate": 2.408952959028832e-06,
"loss": 0.5023,
"step": 635
},
{
"epoch": 5.25,
"learning_rate": 2.427921092564492e-06,
"loss": 0.5426,
"step": 640
},
{
"epoch": 5.29,
"learning_rate": 2.446889226100152e-06,
"loss": 0.4777,
"step": 645
},
{
"epoch": 5.33,
"learning_rate": 2.465857359635812e-06,
"loss": 0.5568,
"step": 650
},
{
"epoch": 5.37,
"learning_rate": 2.484825493171472e-06,
"loss": 0.4905,
"step": 655
},
{
"epoch": 5.41,
"learning_rate": 2.5037936267071324e-06,
"loss": 0.5814,
"step": 660
},
{
"epoch": 5.45,
"learning_rate": 2.5227617602427923e-06,
"loss": 0.5155,
"step": 665
},
{
"epoch": 5.49,
"learning_rate": 2.5417298937784526e-06,
"loss": 0.4845,
"step": 670
},
{
"epoch": 5.53,
"learning_rate": 2.5606980273141125e-06,
"loss": 0.4075,
"step": 675
},
{
"epoch": 5.57,
"learning_rate": 2.5796661608497724e-06,
"loss": 0.4734,
"step": 680
},
{
"epoch": 5.61,
"learning_rate": 2.5986342943854327e-06,
"loss": 0.4504,
"step": 685
},
{
"epoch": 5.66,
"learning_rate": 2.6176024279210926e-06,
"loss": 0.507,
"step": 690
},
{
"epoch": 5.7,
"learning_rate": 2.636570561456753e-06,
"loss": 0.4954,
"step": 695
},
{
"epoch": 5.74,
"learning_rate": 2.655538694992413e-06,
"loss": 0.4318,
"step": 700
},
{
"epoch": 5.78,
"learning_rate": 2.6745068285280727e-06,
"loss": 0.534,
"step": 705
},
{
"epoch": 5.82,
"learning_rate": 2.693474962063733e-06,
"loss": 0.5475,
"step": 710
},
{
"epoch": 5.86,
"learning_rate": 2.712443095599393e-06,
"loss": 0.4276,
"step": 715
},
{
"epoch": 5.9,
"learning_rate": 2.7314112291350532e-06,
"loss": 0.4613,
"step": 720
},
{
"epoch": 5.94,
"learning_rate": 2.7503793626707136e-06,
"loss": 0.4466,
"step": 725
},
{
"epoch": 5.98,
"learning_rate": 2.769347496206374e-06,
"loss": 0.5246,
"step": 730
},
{
"epoch": 6.0,
"eval_accuracy": 0.8735868448098664,
"eval_loss": 0.35859864950180054,
"eval_runtime": 38.3347,
"eval_samples_per_second": 25.382,
"eval_steps_per_second": 1.591,
"step": 732
},
{
"epoch": 6.02,
"learning_rate": 2.7883156297420338e-06,
"loss": 0.4048,
"step": 735
},
{
"epoch": 6.07,
"learning_rate": 2.807283763277694e-06,
"loss": 0.5199,
"step": 740
},
{
"epoch": 6.11,
"learning_rate": 2.826251896813354e-06,
"loss": 0.4736,
"step": 745
},
{
"epoch": 6.15,
"learning_rate": 2.845220030349014e-06,
"loss": 0.4186,
"step": 750
},
{
"epoch": 6.19,
"learning_rate": 2.8641881638846742e-06,
"loss": 0.4876,
"step": 755
},
{
"epoch": 6.23,
"learning_rate": 2.883156297420334e-06,
"loss": 0.4284,
"step": 760
},
{
"epoch": 6.27,
"learning_rate": 2.9021244309559944e-06,
"loss": 0.3561,
"step": 765
},
{
"epoch": 6.31,
"learning_rate": 2.9210925644916543e-06,
"loss": 0.4445,
"step": 770
},
{
"epoch": 6.35,
"learning_rate": 2.9400606980273142e-06,
"loss": 0.4215,
"step": 775
},
{
"epoch": 6.39,
"learning_rate": 2.9590288315629746e-06,
"loss": 0.4689,
"step": 780
},
{
"epoch": 6.43,
"learning_rate": 2.9779969650986345e-06,
"loss": 0.4195,
"step": 785
},
{
"epoch": 6.48,
"learning_rate": 2.9969650986342948e-06,
"loss": 0.4175,
"step": 790
},
{
"epoch": 6.52,
"learning_rate": 3.0159332321699547e-06,
"loss": 0.4197,
"step": 795
},
{
"epoch": 6.56,
"learning_rate": 3.0349013657056146e-06,
"loss": 0.3741,
"step": 800
},
{
"epoch": 6.6,
"learning_rate": 3.053869499241275e-06,
"loss": 0.4688,
"step": 805
},
{
"epoch": 6.64,
"learning_rate": 3.072837632776935e-06,
"loss": 0.4878,
"step": 810
},
{
"epoch": 6.68,
"learning_rate": 3.091805766312595e-06,
"loss": 0.4784,
"step": 815
},
{
"epoch": 6.72,
"learning_rate": 3.110773899848255e-06,
"loss": 0.4877,
"step": 820
},
{
"epoch": 6.76,
"learning_rate": 3.1297420333839153e-06,
"loss": 0.5009,
"step": 825
},
{
"epoch": 6.8,
"learning_rate": 3.1487101669195752e-06,
"loss": 0.431,
"step": 830
},
{
"epoch": 6.84,
"learning_rate": 3.167678300455235e-06,
"loss": 0.3757,
"step": 835
},
{
"epoch": 6.89,
"learning_rate": 3.1866464339908955e-06,
"loss": 0.3282,
"step": 840
},
{
"epoch": 6.93,
"learning_rate": 3.2056145675265554e-06,
"loss": 0.3639,
"step": 845
},
{
"epoch": 6.97,
"learning_rate": 3.2245827010622157e-06,
"loss": 0.3691,
"step": 850
},
{
"epoch": 7.0,
"eval_accuracy": 0.8992805755395683,
"eval_loss": 0.31340497732162476,
"eval_runtime": 38.469,
"eval_samples_per_second": 25.293,
"eval_steps_per_second": 1.586,
"step": 854
},
{
"epoch": 7.01,
"learning_rate": 3.2435508345978756e-06,
"loss": 0.3124,
"step": 855
},
{
"epoch": 7.05,
"learning_rate": 3.2625189681335363e-06,
"loss": 0.3577,
"step": 860
},
{
"epoch": 7.09,
"learning_rate": 3.2814871016691962e-06,
"loss": 0.2683,
"step": 865
},
{
"epoch": 7.13,
"learning_rate": 3.3004552352048565e-06,
"loss": 0.2743,
"step": 870
},
{
"epoch": 7.17,
"learning_rate": 3.3194233687405164e-06,
"loss": 0.3645,
"step": 875
},
{
"epoch": 7.21,
"learning_rate": 3.3383915022761763e-06,
"loss": 0.3546,
"step": 880
},
{
"epoch": 7.25,
"learning_rate": 3.3573596358118367e-06,
"loss": 0.4235,
"step": 885
},
{
"epoch": 7.3,
"learning_rate": 3.3763277693474965e-06,
"loss": 0.3837,
"step": 890
},
{
"epoch": 7.34,
"learning_rate": 3.395295902883157e-06,
"loss": 0.2884,
"step": 895
},
{
"epoch": 7.38,
"learning_rate": 3.4142640364188168e-06,
"loss": 0.3577,
"step": 900
},
{
"epoch": 7.42,
"learning_rate": 3.4332321699544767e-06,
"loss": 0.3608,
"step": 905
},
{
"epoch": 7.46,
"learning_rate": 3.452200303490137e-06,
"loss": 0.3299,
"step": 910
},
{
"epoch": 7.5,
"learning_rate": 3.471168437025797e-06,
"loss": 0.3111,
"step": 915
},
{
"epoch": 7.54,
"learning_rate": 3.490136570561457e-06,
"loss": 0.3328,
"step": 920
},
{
"epoch": 7.58,
"learning_rate": 3.509104704097117e-06,
"loss": 0.2494,
"step": 925
},
{
"epoch": 7.62,
"learning_rate": 3.528072837632777e-06,
"loss": 0.3438,
"step": 930
},
{
"epoch": 7.66,
"learning_rate": 3.5470409711684373e-06,
"loss": 0.3035,
"step": 935
},
{
"epoch": 7.7,
"learning_rate": 3.5660091047040972e-06,
"loss": 0.2355,
"step": 940
},
{
"epoch": 7.75,
"learning_rate": 3.5849772382397575e-06,
"loss": 0.2861,
"step": 945
},
{
"epoch": 7.79,
"learning_rate": 3.6039453717754174e-06,
"loss": 0.2461,
"step": 950
},
{
"epoch": 7.83,
"learning_rate": 3.6229135053110773e-06,
"loss": 0.3799,
"step": 955
},
{
"epoch": 7.87,
"learning_rate": 3.6418816388467377e-06,
"loss": 0.3158,
"step": 960
},
{
"epoch": 7.91,
"learning_rate": 3.6608497723823976e-06,
"loss": 0.3154,
"step": 965
},
{
"epoch": 7.95,
"learning_rate": 3.679817905918058e-06,
"loss": 0.3094,
"step": 970
},
{
"epoch": 7.99,
"learning_rate": 3.6987860394537178e-06,
"loss": 0.4708,
"step": 975
},
{
"epoch": 8.0,
"eval_accuracy": 0.9393627954779034,
"eval_loss": 0.20690585672855377,
"eval_runtime": 38.3529,
"eval_samples_per_second": 25.37,
"eval_steps_per_second": 1.59,
"step": 976
},
{
"epoch": 8.03,
"learning_rate": 3.717754172989378e-06,
"loss": 0.2706,
"step": 980
},
{
"epoch": 8.07,
"learning_rate": 3.736722306525038e-06,
"loss": 0.2366,
"step": 985
},
{
"epoch": 8.11,
"learning_rate": 3.7556904400606987e-06,
"loss": 0.3565,
"step": 990
},
{
"epoch": 8.16,
"learning_rate": 3.7746585735963586e-06,
"loss": 0.2274,
"step": 995
},
{
"epoch": 8.2,
"learning_rate": 3.7936267071320185e-06,
"loss": 0.2032,
"step": 1000
},
{
"epoch": 8.24,
"learning_rate": 3.812594840667679e-06,
"loss": 0.3676,
"step": 1005
},
{
"epoch": 8.28,
"learning_rate": 3.831562974203339e-06,
"loss": 0.3325,
"step": 1010
},
{
"epoch": 8.32,
"learning_rate": 3.850531107738999e-06,
"loss": 0.2529,
"step": 1015
},
{
"epoch": 8.36,
"learning_rate": 3.869499241274659e-06,
"loss": 0.3142,
"step": 1020
},
{
"epoch": 8.4,
"learning_rate": 3.888467374810319e-06,
"loss": 0.2762,
"step": 1025
},
{
"epoch": 8.44,
"learning_rate": 3.907435508345979e-06,
"loss": 0.307,
"step": 1030
},
{
"epoch": 8.48,
"learning_rate": 3.9264036418816395e-06,
"loss": 0.3052,
"step": 1035
},
{
"epoch": 8.52,
"learning_rate": 3.945371775417299e-06,
"loss": 0.2407,
"step": 1040
},
{
"epoch": 8.57,
"learning_rate": 3.964339908952959e-06,
"loss": 0.3083,
"step": 1045
},
{
"epoch": 8.61,
"learning_rate": 3.983308042488619e-06,
"loss": 0.2535,
"step": 1050
},
{
"epoch": 8.65,
"learning_rate": 4.00227617602428e-06,
"loss": 0.2663,
"step": 1055
},
{
"epoch": 8.69,
"learning_rate": 4.02124430955994e-06,
"loss": 0.29,
"step": 1060
},
{
"epoch": 8.73,
"learning_rate": 4.0402124430956e-06,
"loss": 0.3125,
"step": 1065
},
{
"epoch": 8.77,
"learning_rate": 4.05918057663126e-06,
"loss": 0.3293,
"step": 1070
},
{
"epoch": 8.81,
"learning_rate": 4.0781487101669195e-06,
"loss": 0.2345,
"step": 1075
},
{
"epoch": 8.85,
"learning_rate": 4.09711684370258e-06,
"loss": 0.2147,
"step": 1080
},
{
"epoch": 8.89,
"learning_rate": 4.11608497723824e-06,
"loss": 0.1732,
"step": 1085
},
{
"epoch": 8.93,
"learning_rate": 4.1350531107739e-06,
"loss": 0.2756,
"step": 1090
},
{
"epoch": 8.98,
"learning_rate": 4.15402124430956e-06,
"loss": 0.1694,
"step": 1095
},
{
"epoch": 9.0,
"eval_accuracy": 0.9414182939362795,
"eval_loss": 0.18316909670829773,
"eval_runtime": 38.3501,
"eval_samples_per_second": 25.371,
"eval_steps_per_second": 1.591,
"step": 1098
},
{
"epoch": 9.02,
"learning_rate": 4.17298937784522e-06,
"loss": 0.272,
"step": 1100
},
{
"epoch": 9.06,
"learning_rate": 4.191957511380881e-06,
"loss": 0.2576,
"step": 1105
},
{
"epoch": 9.1,
"learning_rate": 4.2109256449165405e-06,
"loss": 0.2456,
"step": 1110
},
{
"epoch": 9.14,
"learning_rate": 4.2298937784522004e-06,
"loss": 0.2365,
"step": 1115
},
{
"epoch": 9.18,
"learning_rate": 4.24886191198786e-06,
"loss": 0.1883,
"step": 1120
},
{
"epoch": 9.22,
"learning_rate": 4.267830045523521e-06,
"loss": 0.2595,
"step": 1125
},
{
"epoch": 9.26,
"learning_rate": 4.286798179059181e-06,
"loss": 0.1742,
"step": 1130
},
{
"epoch": 9.3,
"learning_rate": 4.305766312594841e-06,
"loss": 0.2042,
"step": 1135
},
{
"epoch": 9.34,
"learning_rate": 4.324734446130502e-06,
"loss": 0.1657,
"step": 1140
},
{
"epoch": 9.39,
"learning_rate": 4.3437025796661615e-06,
"loss": 0.2747,
"step": 1145
},
{
"epoch": 9.43,
"learning_rate": 4.362670713201821e-06,
"loss": 0.1626,
"step": 1150
},
{
"epoch": 9.47,
"learning_rate": 4.381638846737481e-06,
"loss": 0.2099,
"step": 1155
},
{
"epoch": 9.51,
"learning_rate": 4.400606980273141e-06,
"loss": 0.1882,
"step": 1160
},
{
"epoch": 9.55,
"learning_rate": 4.419575113808802e-06,
"loss": 0.3238,
"step": 1165
},
{
"epoch": 9.59,
"learning_rate": 4.438543247344462e-06,
"loss": 0.1972,
"step": 1170
},
{
"epoch": 9.63,
"learning_rate": 4.457511380880122e-06,
"loss": 0.2717,
"step": 1175
},
{
"epoch": 9.67,
"learning_rate": 4.476479514415782e-06,
"loss": 0.3135,
"step": 1180
},
{
"epoch": 9.71,
"learning_rate": 4.4954476479514415e-06,
"loss": 0.2084,
"step": 1185
},
{
"epoch": 9.75,
"learning_rate": 4.514415781487102e-06,
"loss": 0.207,
"step": 1190
},
{
"epoch": 9.8,
"learning_rate": 4.533383915022762e-06,
"loss": 0.1648,
"step": 1195
},
{
"epoch": 9.84,
"learning_rate": 4.552352048558422e-06,
"loss": 0.2692,
"step": 1200
},
{
"epoch": 9.88,
"learning_rate": 4.571320182094082e-06,
"loss": 0.2485,
"step": 1205
},
{
"epoch": 9.92,
"learning_rate": 4.590288315629743e-06,
"loss": 0.2064,
"step": 1210
},
{
"epoch": 9.96,
"learning_rate": 4.609256449165403e-06,
"loss": 0.1995,
"step": 1215
},
{
"epoch": 10.0,
"learning_rate": 4.6282245827010625e-06,
"loss": 0.2749,
"step": 1220
},
{
"epoch": 10.0,
"eval_accuracy": 0.9640287769784173,
"eval_loss": 0.11980544030666351,
"eval_runtime": 38.2362,
"eval_samples_per_second": 25.447,
"eval_steps_per_second": 1.595,
"step": 1220
},
{
"epoch": 10.04,
"learning_rate": 4.647192716236722e-06,
"loss": 0.2508,
"step": 1225
},
{
"epoch": 10.08,
"learning_rate": 4.666160849772382e-06,
"loss": 0.1797,
"step": 1230
},
{
"epoch": 10.12,
"learning_rate": 4.685128983308043e-06,
"loss": 0.2694,
"step": 1235
},
{
"epoch": 10.16,
"learning_rate": 4.704097116843703e-06,
"loss": 0.2526,
"step": 1240
},
{
"epoch": 10.2,
"learning_rate": 4.723065250379363e-06,
"loss": 0.2319,
"step": 1245
},
{
"epoch": 10.25,
"learning_rate": 4.742033383915023e-06,
"loss": 0.2352,
"step": 1250
},
{
"epoch": 10.29,
"learning_rate": 4.7610015174506835e-06,
"loss": 0.263,
"step": 1255
},
{
"epoch": 10.33,
"learning_rate": 4.779969650986343e-06,
"loss": 0.1787,
"step": 1260
},
{
"epoch": 10.37,
"learning_rate": 4.798937784522003e-06,
"loss": 0.1999,
"step": 1265
},
{
"epoch": 10.41,
"learning_rate": 4.817905918057664e-06,
"loss": 0.2165,
"step": 1270
},
{
"epoch": 10.45,
"learning_rate": 4.836874051593324e-06,
"loss": 0.1601,
"step": 1275
},
{
"epoch": 10.49,
"learning_rate": 4.855842185128984e-06,
"loss": 0.1765,
"step": 1280
},
{
"epoch": 10.53,
"learning_rate": 4.874810318664644e-06,
"loss": 0.196,
"step": 1285
},
{
"epoch": 10.57,
"learning_rate": 4.893778452200304e-06,
"loss": 0.1229,
"step": 1290
},
{
"epoch": 10.61,
"learning_rate": 4.912746585735964e-06,
"loss": 0.1257,
"step": 1295
},
{
"epoch": 10.66,
"learning_rate": 4.931714719271624e-06,
"loss": 0.1496,
"step": 1300
},
{
"epoch": 10.7,
"learning_rate": 4.950682852807284e-06,
"loss": 0.17,
"step": 1305
},
{
"epoch": 10.74,
"learning_rate": 4.969650986342944e-06,
"loss": 0.1545,
"step": 1310
},
{
"epoch": 10.78,
"learning_rate": 4.988619119878604e-06,
"loss": 0.1693,
"step": 1315
},
{
"epoch": 10.82,
"learning_rate": 4.931506849315069e-06,
"loss": 0.1291,
"step": 1320
},
{
"epoch": 10.86,
"learning_rate": 4.76027397260274e-06,
"loss": 0.1815,
"step": 1325
},
{
"epoch": 10.9,
"learning_rate": 4.589041095890411e-06,
"loss": 0.1872,
"step": 1330
},
{
"epoch": 10.94,
"learning_rate": 4.4178082191780825e-06,
"loss": 0.2,
"step": 1335
},
{
"epoch": 10.98,
"learning_rate": 4.246575342465754e-06,
"loss": 0.1777,
"step": 1340
},
{
"epoch": 11.0,
"eval_accuracy": 0.9732785200411099,
"eval_loss": 0.0844743624329567,
"eval_runtime": 38.4074,
"eval_samples_per_second": 25.334,
"eval_steps_per_second": 1.588,
"step": 1342
},
{
"epoch": 11.02,
"learning_rate": 4.075342465753426e-06,
"loss": 0.2062,
"step": 1345
},
{
"epoch": 11.07,
"learning_rate": 3.904109589041096e-06,
"loss": 0.1519,
"step": 1350
},
{
"epoch": 11.11,
"learning_rate": 3.7328767123287675e-06,
"loss": 0.1812,
"step": 1355
},
{
"epoch": 11.15,
"learning_rate": 3.5616438356164386e-06,
"loss": 0.1496,
"step": 1360
},
{
"epoch": 11.19,
"learning_rate": 3.39041095890411e-06,
"loss": 0.1927,
"step": 1365
},
{
"epoch": 11.23,
"learning_rate": 3.2191780821917813e-06,
"loss": 0.151,
"step": 1370
},
{
"epoch": 11.27,
"learning_rate": 3.0479452054794525e-06,
"loss": 0.1262,
"step": 1375
},
{
"epoch": 11.31,
"learning_rate": 2.876712328767123e-06,
"loss": 0.2034,
"step": 1380
},
{
"epoch": 11.35,
"learning_rate": 2.705479452054795e-06,
"loss": 0.1513,
"step": 1385
},
{
"epoch": 11.39,
"learning_rate": 2.534246575342466e-06,
"loss": 0.1479,
"step": 1390
},
{
"epoch": 11.43,
"learning_rate": 2.363013698630137e-06,
"loss": 0.0822,
"step": 1395
},
{
"epoch": 11.48,
"learning_rate": 2.191780821917808e-06,
"loss": 0.1607,
"step": 1400
},
{
"epoch": 11.52,
"learning_rate": 2.0205479452054797e-06,
"loss": 0.0946,
"step": 1405
},
{
"epoch": 11.56,
"learning_rate": 1.8493150684931507e-06,
"loss": 0.1685,
"step": 1410
},
{
"epoch": 11.6,
"learning_rate": 1.678082191780822e-06,
"loss": 0.1005,
"step": 1415
},
{
"epoch": 11.64,
"learning_rate": 1.5068493150684932e-06,
"loss": 0.1871,
"step": 1420
},
{
"epoch": 11.68,
"learning_rate": 1.3356164383561645e-06,
"loss": 0.1389,
"step": 1425
},
{
"epoch": 11.72,
"learning_rate": 1.1643835616438357e-06,
"loss": 0.1835,
"step": 1430
},
{
"epoch": 11.76,
"learning_rate": 9.931506849315068e-07,
"loss": 0.1698,
"step": 1435
},
{
"epoch": 11.8,
"learning_rate": 8.219178082191781e-07,
"loss": 0.2009,
"step": 1440
},
{
"epoch": 11.84,
"learning_rate": 6.506849315068493e-07,
"loss": 0.1655,
"step": 1445
},
{
"epoch": 11.89,
"learning_rate": 4.794520547945206e-07,
"loss": 0.1749,
"step": 1450
},
{
"epoch": 11.93,
"learning_rate": 3.082191780821918e-07,
"loss": 0.1388,
"step": 1455
},
{
"epoch": 11.97,
"learning_rate": 1.36986301369863e-07,
"loss": 0.1529,
"step": 1460
},
{
"epoch": 12.0,
"eval_accuracy": 0.9907502569373073,
"eval_loss": 0.04338991269469261,
"eval_runtime": 38.35,
"eval_samples_per_second": 25.372,
"eval_steps_per_second": 1.591,
"step": 1464
},
{
"epoch": 12.0,
"step": 1464,
"total_flos": 1.2795133907158622e+19,
"train_loss": 0.7126436810024449,
"train_runtime": 8746.6735,
"train_samples_per_second": 5.338,
"train_steps_per_second": 0.167
}
],
"max_steps": 1464,
"num_train_epochs": 12,
"total_flos": 1.2795133907158622e+19,
"trial_name": null,
"trial_params": null
}