ALM-AHME's picture
End of training
1e8c96f
{
"best_metric": 0.986639260020555,
"best_model_checkpoint": "convnextv2-large-1k-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": 1.8968133535660094e-07,
"loss": 1.9517,
"step": 5
},
{
"epoch": 0.08,
"learning_rate": 3.793626707132019e-07,
"loss": 1.9476,
"step": 10
},
{
"epoch": 0.12,
"learning_rate": 5.690440060698028e-07,
"loss": 1.9461,
"step": 15
},
{
"epoch": 0.16,
"learning_rate": 7.587253414264037e-07,
"loss": 1.9429,
"step": 20
},
{
"epoch": 0.2,
"learning_rate": 9.484066767830046e-07,
"loss": 1.9471,
"step": 25
},
{
"epoch": 0.25,
"learning_rate": 1.1380880121396055e-06,
"loss": 1.9423,
"step": 30
},
{
"epoch": 0.29,
"learning_rate": 1.3277693474962064e-06,
"loss": 1.9447,
"step": 35
},
{
"epoch": 0.33,
"learning_rate": 1.5174506828528075e-06,
"loss": 1.9379,
"step": 40
},
{
"epoch": 0.37,
"learning_rate": 1.7071320182094082e-06,
"loss": 1.9426,
"step": 45
},
{
"epoch": 0.41,
"learning_rate": 1.8968133535660093e-06,
"loss": 1.9459,
"step": 50
},
{
"epoch": 0.45,
"learning_rate": 2.0864946889226104e-06,
"loss": 1.9322,
"step": 55
},
{
"epoch": 0.49,
"learning_rate": 2.276176024279211e-06,
"loss": 1.9376,
"step": 60
},
{
"epoch": 0.53,
"learning_rate": 2.465857359635812e-06,
"loss": 1.9398,
"step": 65
},
{
"epoch": 0.57,
"learning_rate": 2.655538694992413e-06,
"loss": 1.9288,
"step": 70
},
{
"epoch": 0.61,
"learning_rate": 2.845220030349014e-06,
"loss": 1.9343,
"step": 75
},
{
"epoch": 0.66,
"learning_rate": 3.034901365705615e-06,
"loss": 1.9235,
"step": 80
},
{
"epoch": 0.7,
"learning_rate": 3.2245827010622157e-06,
"loss": 1.9326,
"step": 85
},
{
"epoch": 0.74,
"learning_rate": 3.4142640364188163e-06,
"loss": 1.9155,
"step": 90
},
{
"epoch": 0.78,
"learning_rate": 3.6039453717754174e-06,
"loss": 1.9093,
"step": 95
},
{
"epoch": 0.82,
"learning_rate": 3.7936267071320185e-06,
"loss": 1.9142,
"step": 100
},
{
"epoch": 0.86,
"learning_rate": 3.983308042488619e-06,
"loss": 1.9145,
"step": 105
},
{
"epoch": 0.9,
"learning_rate": 4.172989377845221e-06,
"loss": 1.9174,
"step": 110
},
{
"epoch": 0.94,
"learning_rate": 4.362670713201821e-06,
"loss": 1.9059,
"step": 115
},
{
"epoch": 0.98,
"learning_rate": 4.552352048558422e-06,
"loss": 1.8977,
"step": 120
},
{
"epoch": 1.0,
"eval_accuracy": 0.29393627954779034,
"eval_loss": 1.8948885202407837,
"eval_runtime": 584.5436,
"eval_samples_per_second": 1.665,
"eval_steps_per_second": 0.104,
"step": 122
},
{
"epoch": 1.02,
"learning_rate": 4.742033383915024e-06,
"loss": 1.8998,
"step": 125
},
{
"epoch": 1.07,
"learning_rate": 4.931714719271624e-06,
"loss": 1.8885,
"step": 130
},
{
"epoch": 1.11,
"learning_rate": 5.121396054628225e-06,
"loss": 1.8772,
"step": 135
},
{
"epoch": 1.15,
"learning_rate": 5.311077389984826e-06,
"loss": 1.8688,
"step": 140
},
{
"epoch": 1.19,
"learning_rate": 5.500758725341427e-06,
"loss": 1.8569,
"step": 145
},
{
"epoch": 1.23,
"learning_rate": 5.690440060698028e-06,
"loss": 1.8428,
"step": 150
},
{
"epoch": 1.27,
"learning_rate": 5.8801213960546285e-06,
"loss": 1.8231,
"step": 155
},
{
"epoch": 1.31,
"learning_rate": 6.06980273141123e-06,
"loss": 1.8507,
"step": 160
},
{
"epoch": 1.35,
"learning_rate": 6.259484066767831e-06,
"loss": 1.821,
"step": 165
},
{
"epoch": 1.39,
"learning_rate": 6.449165402124431e-06,
"loss": 1.8163,
"step": 170
},
{
"epoch": 1.43,
"learning_rate": 6.638846737481032e-06,
"loss": 1.8093,
"step": 175
},
{
"epoch": 1.48,
"learning_rate": 6.828528072837633e-06,
"loss": 1.8103,
"step": 180
},
{
"epoch": 1.52,
"learning_rate": 7.018209408194234e-06,
"loss": 1.7829,
"step": 185
},
{
"epoch": 1.56,
"learning_rate": 7.207890743550835e-06,
"loss": 1.7511,
"step": 190
},
{
"epoch": 1.6,
"learning_rate": 7.3975720789074356e-06,
"loss": 1.7638,
"step": 195
},
{
"epoch": 1.64,
"learning_rate": 7.587253414264037e-06,
"loss": 1.7292,
"step": 200
},
{
"epoch": 1.68,
"learning_rate": 7.776934749620638e-06,
"loss": 1.7184,
"step": 205
},
{
"epoch": 1.72,
"learning_rate": 7.966616084977238e-06,
"loss": 1.6952,
"step": 210
},
{
"epoch": 1.76,
"learning_rate": 8.15629742033384e-06,
"loss": 1.7019,
"step": 215
},
{
"epoch": 1.8,
"learning_rate": 8.345978755690441e-06,
"loss": 1.6731,
"step": 220
},
{
"epoch": 1.84,
"learning_rate": 8.535660091047042e-06,
"loss": 1.6475,
"step": 225
},
{
"epoch": 1.89,
"learning_rate": 8.725341426403643e-06,
"loss": 1.6799,
"step": 230
},
{
"epoch": 1.93,
"learning_rate": 8.915022761760243e-06,
"loss": 1.6905,
"step": 235
},
{
"epoch": 1.97,
"learning_rate": 9.104704097116844e-06,
"loss": 1.6493,
"step": 240
},
{
"epoch": 2.0,
"eval_accuracy": 0.5447070914696814,
"eval_loss": 1.64486563205719,
"eval_runtime": 12.1588,
"eval_samples_per_second": 80.024,
"eval_steps_per_second": 5.017,
"step": 244
},
{
"epoch": 2.01,
"learning_rate": 9.294385432473445e-06,
"loss": 1.627,
"step": 245
},
{
"epoch": 2.05,
"learning_rate": 9.484066767830047e-06,
"loss": 1.6143,
"step": 250
},
{
"epoch": 2.09,
"learning_rate": 9.673748103186648e-06,
"loss": 1.6195,
"step": 255
},
{
"epoch": 2.13,
"learning_rate": 9.863429438543249e-06,
"loss": 1.6314,
"step": 260
},
{
"epoch": 2.17,
"learning_rate": 1.005311077389985e-05,
"loss": 1.5916,
"step": 265
},
{
"epoch": 2.21,
"learning_rate": 1.024279210925645e-05,
"loss": 1.6053,
"step": 270
},
{
"epoch": 2.25,
"learning_rate": 1.043247344461305e-05,
"loss": 1.581,
"step": 275
},
{
"epoch": 2.3,
"learning_rate": 1.0622154779969651e-05,
"loss": 1.5884,
"step": 280
},
{
"epoch": 2.34,
"learning_rate": 1.0811836115326254e-05,
"loss": 1.5753,
"step": 285
},
{
"epoch": 2.38,
"learning_rate": 1.1001517450682854e-05,
"loss": 1.5267,
"step": 290
},
{
"epoch": 2.42,
"learning_rate": 1.1191198786039455e-05,
"loss": 1.5596,
"step": 295
},
{
"epoch": 2.46,
"learning_rate": 1.1380880121396056e-05,
"loss": 1.5722,
"step": 300
},
{
"epoch": 2.5,
"learning_rate": 1.1570561456752656e-05,
"loss": 1.5245,
"step": 305
},
{
"epoch": 2.54,
"learning_rate": 1.1760242792109257e-05,
"loss": 1.4489,
"step": 310
},
{
"epoch": 2.58,
"learning_rate": 1.194992412746586e-05,
"loss": 1.4772,
"step": 315
},
{
"epoch": 2.62,
"learning_rate": 1.213960546282246e-05,
"loss": 1.4503,
"step": 320
},
{
"epoch": 2.66,
"learning_rate": 1.232928679817906e-05,
"loss": 1.4399,
"step": 325
},
{
"epoch": 2.7,
"learning_rate": 1.2518968133535661e-05,
"loss": 1.3995,
"step": 330
},
{
"epoch": 2.75,
"learning_rate": 1.2708649468892262e-05,
"loss": 1.3476,
"step": 335
},
{
"epoch": 2.79,
"learning_rate": 1.2898330804248863e-05,
"loss": 1.341,
"step": 340
},
{
"epoch": 2.83,
"learning_rate": 1.3088012139605463e-05,
"loss": 1.3717,
"step": 345
},
{
"epoch": 2.87,
"learning_rate": 1.3277693474962064e-05,
"loss": 1.3892,
"step": 350
},
{
"epoch": 2.91,
"learning_rate": 1.3467374810318665e-05,
"loss": 1.376,
"step": 355
},
{
"epoch": 2.95,
"learning_rate": 1.3657056145675265e-05,
"loss": 1.2725,
"step": 360
},
{
"epoch": 2.99,
"learning_rate": 1.3846737481031868e-05,
"loss": 1.239,
"step": 365
},
{
"epoch": 3.0,
"eval_accuracy": 0.6885919835560124,
"eval_loss": 1.2818713188171387,
"eval_runtime": 12.0408,
"eval_samples_per_second": 80.808,
"eval_steps_per_second": 5.066,
"step": 366
},
{
"epoch": 3.03,
"learning_rate": 1.4036418816388468e-05,
"loss": 1.3048,
"step": 370
},
{
"epoch": 3.07,
"learning_rate": 1.4226100151745069e-05,
"loss": 1.2479,
"step": 375
},
{
"epoch": 3.11,
"learning_rate": 1.441578148710167e-05,
"loss": 1.2245,
"step": 380
},
{
"epoch": 3.16,
"learning_rate": 1.460546282245827e-05,
"loss": 1.1883,
"step": 385
},
{
"epoch": 3.2,
"learning_rate": 1.4795144157814871e-05,
"loss": 1.2198,
"step": 390
},
{
"epoch": 3.24,
"learning_rate": 1.4984825493171473e-05,
"loss": 1.2452,
"step": 395
},
{
"epoch": 3.28,
"learning_rate": 1.5174506828528074e-05,
"loss": 1.1089,
"step": 400
},
{
"epoch": 3.32,
"learning_rate": 1.5364188163884675e-05,
"loss": 1.2382,
"step": 405
},
{
"epoch": 3.36,
"learning_rate": 1.5553869499241275e-05,
"loss": 1.1713,
"step": 410
},
{
"epoch": 3.4,
"learning_rate": 1.5743550834597876e-05,
"loss": 1.1526,
"step": 415
},
{
"epoch": 3.44,
"learning_rate": 1.5933232169954477e-05,
"loss": 1.0863,
"step": 420
},
{
"epoch": 3.48,
"learning_rate": 1.6122913505311078e-05,
"loss": 1.1086,
"step": 425
},
{
"epoch": 3.52,
"learning_rate": 1.631259484066768e-05,
"loss": 1.0015,
"step": 430
},
{
"epoch": 3.57,
"learning_rate": 1.6502276176024282e-05,
"loss": 1.1095,
"step": 435
},
{
"epoch": 3.61,
"learning_rate": 1.6691957511380883e-05,
"loss": 1.1351,
"step": 440
},
{
"epoch": 3.65,
"learning_rate": 1.6881638846737484e-05,
"loss": 1.1058,
"step": 445
},
{
"epoch": 3.69,
"learning_rate": 1.7071320182094084e-05,
"loss": 1.1137,
"step": 450
},
{
"epoch": 3.73,
"learning_rate": 1.7261001517450685e-05,
"loss": 1.1053,
"step": 455
},
{
"epoch": 3.77,
"learning_rate": 1.7450682852807286e-05,
"loss": 1.0913,
"step": 460
},
{
"epoch": 3.81,
"learning_rate": 1.7640364188163886e-05,
"loss": 0.9692,
"step": 465
},
{
"epoch": 3.85,
"learning_rate": 1.7830045523520487e-05,
"loss": 1.0102,
"step": 470
},
{
"epoch": 3.89,
"learning_rate": 1.8019726858877088e-05,
"loss": 0.9957,
"step": 475
},
{
"epoch": 3.93,
"learning_rate": 1.8209408194233688e-05,
"loss": 0.9606,
"step": 480
},
{
"epoch": 3.98,
"learning_rate": 1.839908952959029e-05,
"loss": 0.9342,
"step": 485
},
{
"epoch": 4.0,
"eval_accuracy": 0.7276464542651593,
"eval_loss": 0.9664044380187988,
"eval_runtime": 12.1112,
"eval_samples_per_second": 80.339,
"eval_steps_per_second": 5.037,
"step": 488
},
{
"epoch": 4.02,
"learning_rate": 1.858877086494689e-05,
"loss": 1.0937,
"step": 490
},
{
"epoch": 4.06,
"learning_rate": 1.8778452200303494e-05,
"loss": 0.9113,
"step": 495
},
{
"epoch": 4.1,
"learning_rate": 1.8968133535660094e-05,
"loss": 0.8594,
"step": 500
},
{
"epoch": 4.14,
"learning_rate": 1.9157814871016695e-05,
"loss": 0.9149,
"step": 505
},
{
"epoch": 4.18,
"learning_rate": 1.9347496206373296e-05,
"loss": 0.882,
"step": 510
},
{
"epoch": 4.22,
"learning_rate": 1.9537177541729896e-05,
"loss": 0.8764,
"step": 515
},
{
"epoch": 4.26,
"learning_rate": 1.9726858877086497e-05,
"loss": 0.8376,
"step": 520
},
{
"epoch": 4.3,
"learning_rate": 1.9916540212443098e-05,
"loss": 0.9023,
"step": 525
},
{
"epoch": 4.34,
"learning_rate": 2.01062215477997e-05,
"loss": 1.007,
"step": 530
},
{
"epoch": 4.39,
"learning_rate": 2.02959028831563e-05,
"loss": 0.8895,
"step": 535
},
{
"epoch": 4.43,
"learning_rate": 2.04855842185129e-05,
"loss": 0.945,
"step": 540
},
{
"epoch": 4.47,
"learning_rate": 2.06752655538695e-05,
"loss": 0.9028,
"step": 545
},
{
"epoch": 4.51,
"learning_rate": 2.08649468892261e-05,
"loss": 0.8643,
"step": 550
},
{
"epoch": 4.55,
"learning_rate": 2.1054628224582702e-05,
"loss": 0.9125,
"step": 555
},
{
"epoch": 4.59,
"learning_rate": 2.1244309559939302e-05,
"loss": 0.811,
"step": 560
},
{
"epoch": 4.63,
"learning_rate": 2.1433990895295907e-05,
"loss": 0.8571,
"step": 565
},
{
"epoch": 4.67,
"learning_rate": 2.1623672230652507e-05,
"loss": 0.8408,
"step": 570
},
{
"epoch": 4.71,
"learning_rate": 2.1813353566009108e-05,
"loss": 0.8885,
"step": 575
},
{
"epoch": 4.75,
"learning_rate": 2.200303490136571e-05,
"loss": 0.822,
"step": 580
},
{
"epoch": 4.8,
"learning_rate": 2.219271623672231e-05,
"loss": 0.7378,
"step": 585
},
{
"epoch": 4.84,
"learning_rate": 2.238239757207891e-05,
"loss": 0.7629,
"step": 590
},
{
"epoch": 4.88,
"learning_rate": 2.257207890743551e-05,
"loss": 0.8337,
"step": 595
},
{
"epoch": 4.92,
"learning_rate": 2.276176024279211e-05,
"loss": 0.9015,
"step": 600
},
{
"epoch": 4.96,
"learning_rate": 2.2951441578148712e-05,
"loss": 0.8478,
"step": 605
},
{
"epoch": 5.0,
"learning_rate": 2.3141122913505313e-05,
"loss": 0.7011,
"step": 610
},
{
"epoch": 5.0,
"eval_accuracy": 0.8355601233299075,
"eval_loss": 0.6759570837020874,
"eval_runtime": 12.1497,
"eval_samples_per_second": 80.084,
"eval_steps_per_second": 5.021,
"step": 610
},
{
"epoch": 5.04,
"learning_rate": 2.3330804248861913e-05,
"loss": 0.7822,
"step": 615
},
{
"epoch": 5.08,
"learning_rate": 2.3520485584218514e-05,
"loss": 0.664,
"step": 620
},
{
"epoch": 5.12,
"learning_rate": 2.3710166919575115e-05,
"loss": 0.7834,
"step": 625
},
{
"epoch": 5.16,
"learning_rate": 2.389984825493172e-05,
"loss": 0.7381,
"step": 630
},
{
"epoch": 5.2,
"learning_rate": 2.408952959028832e-05,
"loss": 0.968,
"step": 635
},
{
"epoch": 5.25,
"learning_rate": 2.427921092564492e-05,
"loss": 0.7325,
"step": 640
},
{
"epoch": 5.29,
"learning_rate": 2.446889226100152e-05,
"loss": 0.7525,
"step": 645
},
{
"epoch": 5.33,
"learning_rate": 2.465857359635812e-05,
"loss": 0.7154,
"step": 650
},
{
"epoch": 5.37,
"learning_rate": 2.4848254931714722e-05,
"loss": 0.6069,
"step": 655
},
{
"epoch": 5.41,
"learning_rate": 2.5037936267071323e-05,
"loss": 0.9597,
"step": 660
},
{
"epoch": 5.45,
"learning_rate": 2.5227617602427923e-05,
"loss": 1.0459,
"step": 665
},
{
"epoch": 5.49,
"learning_rate": 2.5417298937784524e-05,
"loss": 0.7455,
"step": 670
},
{
"epoch": 5.53,
"learning_rate": 2.5606980273141125e-05,
"loss": 0.7745,
"step": 675
},
{
"epoch": 5.57,
"learning_rate": 2.5796661608497725e-05,
"loss": 0.7265,
"step": 680
},
{
"epoch": 5.61,
"learning_rate": 2.5986342943854326e-05,
"loss": 0.7923,
"step": 685
},
{
"epoch": 5.66,
"learning_rate": 2.6176024279210927e-05,
"loss": 0.771,
"step": 690
},
{
"epoch": 5.7,
"learning_rate": 2.6365705614567527e-05,
"loss": 0.665,
"step": 695
},
{
"epoch": 5.74,
"learning_rate": 2.6555386949924128e-05,
"loss": 0.6321,
"step": 700
},
{
"epoch": 5.78,
"learning_rate": 2.674506828528073e-05,
"loss": 0.7338,
"step": 705
},
{
"epoch": 5.82,
"learning_rate": 2.693474962063733e-05,
"loss": 0.8274,
"step": 710
},
{
"epoch": 5.86,
"learning_rate": 2.712443095599393e-05,
"loss": 0.664,
"step": 715
},
{
"epoch": 5.9,
"learning_rate": 2.731411229135053e-05,
"loss": 0.7586,
"step": 720
},
{
"epoch": 5.94,
"learning_rate": 2.7503793626707135e-05,
"loss": 0.6957,
"step": 725
},
{
"epoch": 5.98,
"learning_rate": 2.7693474962063735e-05,
"loss": 0.5809,
"step": 730
},
{
"epoch": 6.0,
"eval_accuracy": 0.8468653648509764,
"eval_loss": 0.5792074203491211,
"eval_runtime": 11.9199,
"eval_samples_per_second": 81.628,
"eval_steps_per_second": 5.117,
"step": 732
},
{
"epoch": 6.02,
"learning_rate": 2.7883156297420336e-05,
"loss": 0.6819,
"step": 735
},
{
"epoch": 6.07,
"learning_rate": 2.8072837632776937e-05,
"loss": 0.5695,
"step": 740
},
{
"epoch": 6.11,
"learning_rate": 2.8262518968133538e-05,
"loss": 0.6056,
"step": 745
},
{
"epoch": 6.15,
"learning_rate": 2.8452200303490138e-05,
"loss": 0.7114,
"step": 750
},
{
"epoch": 6.19,
"learning_rate": 2.864188163884674e-05,
"loss": 0.6814,
"step": 755
},
{
"epoch": 6.23,
"learning_rate": 2.883156297420334e-05,
"loss": 0.6709,
"step": 760
},
{
"epoch": 6.27,
"learning_rate": 2.902124430955994e-05,
"loss": 0.5614,
"step": 765
},
{
"epoch": 6.31,
"learning_rate": 2.921092564491654e-05,
"loss": 0.7163,
"step": 770
},
{
"epoch": 6.35,
"learning_rate": 2.940060698027314e-05,
"loss": 0.6908,
"step": 775
},
{
"epoch": 6.39,
"learning_rate": 2.9590288315629742e-05,
"loss": 0.6304,
"step": 780
},
{
"epoch": 6.43,
"learning_rate": 2.9779969650986346e-05,
"loss": 0.5607,
"step": 785
},
{
"epoch": 6.48,
"learning_rate": 2.9969650986342947e-05,
"loss": 0.4848,
"step": 790
},
{
"epoch": 6.52,
"learning_rate": 3.0159332321699548e-05,
"loss": 0.5005,
"step": 795
},
{
"epoch": 6.56,
"learning_rate": 3.0349013657056148e-05,
"loss": 0.6955,
"step": 800
},
{
"epoch": 6.6,
"learning_rate": 3.0538694992412746e-05,
"loss": 0.5024,
"step": 805
},
{
"epoch": 6.64,
"learning_rate": 3.072837632776935e-05,
"loss": 0.592,
"step": 810
},
{
"epoch": 6.68,
"learning_rate": 3.091805766312595e-05,
"loss": 0.682,
"step": 815
},
{
"epoch": 6.72,
"learning_rate": 3.110773899848255e-05,
"loss": 0.5343,
"step": 820
},
{
"epoch": 6.76,
"learning_rate": 3.129742033383915e-05,
"loss": 0.5193,
"step": 825
},
{
"epoch": 6.8,
"learning_rate": 3.148710166919575e-05,
"loss": 0.4789,
"step": 830
},
{
"epoch": 6.84,
"learning_rate": 3.167678300455235e-05,
"loss": 0.5069,
"step": 835
},
{
"epoch": 6.89,
"learning_rate": 3.1866464339908954e-05,
"loss": 0.5803,
"step": 840
},
{
"epoch": 6.93,
"learning_rate": 3.205614567526555e-05,
"loss": 0.5425,
"step": 845
},
{
"epoch": 6.97,
"learning_rate": 3.2245827010622155e-05,
"loss": 0.4846,
"step": 850
},
{
"epoch": 7.0,
"eval_accuracy": 0.8890030832476875,
"eval_loss": 0.4279514253139496,
"eval_runtime": 12.3928,
"eval_samples_per_second": 78.513,
"eval_steps_per_second": 4.922,
"step": 854
},
{
"epoch": 7.01,
"learning_rate": 3.243550834597875e-05,
"loss": 0.5568,
"step": 855
},
{
"epoch": 7.05,
"learning_rate": 3.262518968133536e-05,
"loss": 0.5243,
"step": 860
},
{
"epoch": 7.09,
"learning_rate": 3.281487101669196e-05,
"loss": 0.447,
"step": 865
},
{
"epoch": 7.13,
"learning_rate": 3.3004552352048564e-05,
"loss": 0.5021,
"step": 870
},
{
"epoch": 7.17,
"learning_rate": 3.319423368740516e-05,
"loss": 0.5527,
"step": 875
},
{
"epoch": 7.21,
"learning_rate": 3.3383915022761766e-05,
"loss": 0.5027,
"step": 880
},
{
"epoch": 7.25,
"learning_rate": 3.357359635811836e-05,
"loss": 0.5587,
"step": 885
},
{
"epoch": 7.3,
"learning_rate": 3.376327769347497e-05,
"loss": 0.4831,
"step": 890
},
{
"epoch": 7.34,
"learning_rate": 3.3952959028831564e-05,
"loss": 0.451,
"step": 895
},
{
"epoch": 7.38,
"learning_rate": 3.414264036418817e-05,
"loss": 0.5807,
"step": 900
},
{
"epoch": 7.42,
"learning_rate": 3.4332321699544766e-05,
"loss": 0.511,
"step": 905
},
{
"epoch": 7.46,
"learning_rate": 3.452200303490137e-05,
"loss": 0.4427,
"step": 910
},
{
"epoch": 7.5,
"learning_rate": 3.471168437025797e-05,
"loss": 0.5123,
"step": 915
},
{
"epoch": 7.54,
"learning_rate": 3.490136570561457e-05,
"loss": 0.4353,
"step": 920
},
{
"epoch": 7.58,
"learning_rate": 3.509104704097117e-05,
"loss": 0.4531,
"step": 925
},
{
"epoch": 7.62,
"learning_rate": 3.528072837632777e-05,
"loss": 0.5628,
"step": 930
},
{
"epoch": 7.66,
"learning_rate": 3.547040971168437e-05,
"loss": 0.5291,
"step": 935
},
{
"epoch": 7.7,
"learning_rate": 3.5660091047040974e-05,
"loss": 0.4894,
"step": 940
},
{
"epoch": 7.75,
"learning_rate": 3.584977238239757e-05,
"loss": 0.3667,
"step": 945
},
{
"epoch": 7.79,
"learning_rate": 3.6039453717754175e-05,
"loss": 0.445,
"step": 950
},
{
"epoch": 7.83,
"learning_rate": 3.622913505311077e-05,
"loss": 0.4218,
"step": 955
},
{
"epoch": 7.87,
"learning_rate": 3.6418816388467377e-05,
"loss": 0.3891,
"step": 960
},
{
"epoch": 7.91,
"learning_rate": 3.6608497723823974e-05,
"loss": 0.556,
"step": 965
},
{
"epoch": 7.95,
"learning_rate": 3.679817905918058e-05,
"loss": 0.6518,
"step": 970
},
{
"epoch": 7.99,
"learning_rate": 3.6987860394537175e-05,
"loss": 0.6914,
"step": 975
},
{
"epoch": 8.0,
"eval_accuracy": 0.8848920863309353,
"eval_loss": 0.41210418939590454,
"eval_runtime": 12.2604,
"eval_samples_per_second": 79.361,
"eval_steps_per_second": 4.975,
"step": 976
},
{
"epoch": 8.03,
"learning_rate": 3.717754172989378e-05,
"loss": 0.4396,
"step": 980
},
{
"epoch": 8.07,
"learning_rate": 3.7367223065250377e-05,
"loss": 0.4276,
"step": 985
},
{
"epoch": 8.11,
"learning_rate": 3.755690440060699e-05,
"loss": 0.4773,
"step": 990
},
{
"epoch": 8.16,
"learning_rate": 3.7746585735963585e-05,
"loss": 0.3802,
"step": 995
},
{
"epoch": 8.2,
"learning_rate": 3.793626707132019e-05,
"loss": 0.3709,
"step": 1000
},
{
"epoch": 8.24,
"learning_rate": 3.8125948406676786e-05,
"loss": 0.4427,
"step": 1005
},
{
"epoch": 8.28,
"learning_rate": 3.831562974203339e-05,
"loss": 0.371,
"step": 1010
},
{
"epoch": 8.32,
"learning_rate": 3.850531107738999e-05,
"loss": 0.425,
"step": 1015
},
{
"epoch": 8.36,
"learning_rate": 3.869499241274659e-05,
"loss": 0.4726,
"step": 1020
},
{
"epoch": 8.4,
"learning_rate": 3.888467374810319e-05,
"loss": 0.5321,
"step": 1025
},
{
"epoch": 8.44,
"learning_rate": 3.907435508345979e-05,
"loss": 0.4126,
"step": 1030
},
{
"epoch": 8.48,
"learning_rate": 3.926403641881639e-05,
"loss": 0.3941,
"step": 1035
},
{
"epoch": 8.52,
"learning_rate": 3.9453717754172994e-05,
"loss": 0.526,
"step": 1040
},
{
"epoch": 8.57,
"learning_rate": 3.964339908952959e-05,
"loss": 0.4235,
"step": 1045
},
{
"epoch": 8.61,
"learning_rate": 3.9833080424886195e-05,
"loss": 0.6282,
"step": 1050
},
{
"epoch": 8.65,
"learning_rate": 4.002276176024279e-05,
"loss": 0.626,
"step": 1055
},
{
"epoch": 8.69,
"learning_rate": 4.02124430955994e-05,
"loss": 0.4725,
"step": 1060
},
{
"epoch": 8.73,
"learning_rate": 4.0402124430955994e-05,
"loss": 0.5334,
"step": 1065
},
{
"epoch": 8.77,
"learning_rate": 4.05918057663126e-05,
"loss": 0.4636,
"step": 1070
},
{
"epoch": 8.81,
"learning_rate": 4.0781487101669195e-05,
"loss": 0.4321,
"step": 1075
},
{
"epoch": 8.85,
"learning_rate": 4.09711684370258e-05,
"loss": 0.4062,
"step": 1080
},
{
"epoch": 8.89,
"learning_rate": 4.11608497723824e-05,
"loss": 0.3655,
"step": 1085
},
{
"epoch": 8.93,
"learning_rate": 4.1350531107739e-05,
"loss": 0.4279,
"step": 1090
},
{
"epoch": 8.98,
"learning_rate": 4.15402124430956e-05,
"loss": 0.3815,
"step": 1095
},
{
"epoch": 9.0,
"eval_accuracy": 0.935251798561151,
"eval_loss": 0.27514535188674927,
"eval_runtime": 12.2348,
"eval_samples_per_second": 79.527,
"eval_steps_per_second": 4.986,
"step": 1098
},
{
"epoch": 9.02,
"learning_rate": 4.17298937784522e-05,
"loss": 0.5233,
"step": 1100
},
{
"epoch": 9.06,
"learning_rate": 4.19195751138088e-05,
"loss": 0.3652,
"step": 1105
},
{
"epoch": 9.1,
"learning_rate": 4.2109256449165404e-05,
"loss": 0.358,
"step": 1110
},
{
"epoch": 9.14,
"learning_rate": 4.2298937784522e-05,
"loss": 0.577,
"step": 1115
},
{
"epoch": 9.18,
"learning_rate": 4.2488619119878605e-05,
"loss": 0.4864,
"step": 1120
},
{
"epoch": 9.22,
"learning_rate": 4.267830045523521e-05,
"loss": 0.4892,
"step": 1125
},
{
"epoch": 9.26,
"learning_rate": 4.286798179059181e-05,
"loss": 0.3746,
"step": 1130
},
{
"epoch": 9.3,
"learning_rate": 4.305766312594841e-05,
"loss": 0.3165,
"step": 1135
},
{
"epoch": 9.34,
"learning_rate": 4.3247344461305014e-05,
"loss": 0.3624,
"step": 1140
},
{
"epoch": 9.39,
"learning_rate": 4.343702579666161e-05,
"loss": 0.2702,
"step": 1145
},
{
"epoch": 9.43,
"learning_rate": 4.3626707132018216e-05,
"loss": 0.3254,
"step": 1150
},
{
"epoch": 9.47,
"learning_rate": 4.381638846737481e-05,
"loss": 0.4096,
"step": 1155
},
{
"epoch": 9.51,
"learning_rate": 4.400606980273142e-05,
"loss": 0.3893,
"step": 1160
},
{
"epoch": 9.55,
"learning_rate": 4.4195751138088014e-05,
"loss": 0.3391,
"step": 1165
},
{
"epoch": 9.59,
"learning_rate": 4.438543247344462e-05,
"loss": 0.3867,
"step": 1170
},
{
"epoch": 9.63,
"learning_rate": 4.4575113808801216e-05,
"loss": 0.4551,
"step": 1175
},
{
"epoch": 9.67,
"learning_rate": 4.476479514415782e-05,
"loss": 0.5414,
"step": 1180
},
{
"epoch": 9.71,
"learning_rate": 4.495447647951442e-05,
"loss": 0.4011,
"step": 1185
},
{
"epoch": 9.75,
"learning_rate": 4.514415781487102e-05,
"loss": 0.3802,
"step": 1190
},
{
"epoch": 9.8,
"learning_rate": 4.533383915022762e-05,
"loss": 0.3787,
"step": 1195
},
{
"epoch": 9.84,
"learning_rate": 4.552352048558422e-05,
"loss": 0.646,
"step": 1200
},
{
"epoch": 9.88,
"learning_rate": 4.571320182094082e-05,
"loss": 0.4731,
"step": 1205
},
{
"epoch": 9.92,
"learning_rate": 4.5902883156297424e-05,
"loss": 0.379,
"step": 1210
},
{
"epoch": 9.96,
"learning_rate": 4.609256449165402e-05,
"loss": 0.3712,
"step": 1215
},
{
"epoch": 10.0,
"learning_rate": 4.6282245827010625e-05,
"loss": 0.2931,
"step": 1220
},
{
"epoch": 10.0,
"eval_accuracy": 0.9198355601233299,
"eval_loss": 0.29800403118133545,
"eval_runtime": 12.1927,
"eval_samples_per_second": 79.802,
"eval_steps_per_second": 5.003,
"step": 1220
},
{
"epoch": 10.04,
"learning_rate": 4.647192716236722e-05,
"loss": 0.3946,
"step": 1225
},
{
"epoch": 10.08,
"learning_rate": 4.6661608497723826e-05,
"loss": 0.4123,
"step": 1230
},
{
"epoch": 10.12,
"learning_rate": 4.6851289833080424e-05,
"loss": 0.2879,
"step": 1235
},
{
"epoch": 10.16,
"learning_rate": 4.704097116843703e-05,
"loss": 0.2714,
"step": 1240
},
{
"epoch": 10.2,
"learning_rate": 4.7230652503793625e-05,
"loss": 0.3913,
"step": 1245
},
{
"epoch": 10.25,
"learning_rate": 4.742033383915023e-05,
"loss": 0.3121,
"step": 1250
},
{
"epoch": 10.29,
"learning_rate": 4.761001517450683e-05,
"loss": 0.4599,
"step": 1255
},
{
"epoch": 10.33,
"learning_rate": 4.779969650986344e-05,
"loss": 0.3933,
"step": 1260
},
{
"epoch": 10.37,
"learning_rate": 4.7989377845220035e-05,
"loss": 0.4022,
"step": 1265
},
{
"epoch": 10.41,
"learning_rate": 4.817905918057664e-05,
"loss": 0.3468,
"step": 1270
},
{
"epoch": 10.45,
"learning_rate": 4.8368740515933236e-05,
"loss": 0.2745,
"step": 1275
},
{
"epoch": 10.49,
"learning_rate": 4.855842185128984e-05,
"loss": 0.3175,
"step": 1280
},
{
"epoch": 10.53,
"learning_rate": 4.874810318664644e-05,
"loss": 0.3136,
"step": 1285
},
{
"epoch": 10.57,
"learning_rate": 4.893778452200304e-05,
"loss": 0.3038,
"step": 1290
},
{
"epoch": 10.61,
"learning_rate": 4.912746585735964e-05,
"loss": 0.1937,
"step": 1295
},
{
"epoch": 10.66,
"learning_rate": 4.931714719271624e-05,
"loss": 0.3836,
"step": 1300
},
{
"epoch": 10.7,
"learning_rate": 4.950682852807284e-05,
"loss": 0.3209,
"step": 1305
},
{
"epoch": 10.74,
"learning_rate": 4.9696509863429444e-05,
"loss": 0.2956,
"step": 1310
},
{
"epoch": 10.78,
"learning_rate": 4.988619119878604e-05,
"loss": 0.3009,
"step": 1315
},
{
"epoch": 10.82,
"learning_rate": 4.9315068493150684e-05,
"loss": 0.2476,
"step": 1320
},
{
"epoch": 10.86,
"learning_rate": 4.7602739726027403e-05,
"loss": 0.3736,
"step": 1325
},
{
"epoch": 10.9,
"learning_rate": 4.589041095890411e-05,
"loss": 0.3156,
"step": 1330
},
{
"epoch": 10.94,
"learning_rate": 4.417808219178082e-05,
"loss": 0.3551,
"step": 1335
},
{
"epoch": 10.98,
"learning_rate": 4.2465753424657536e-05,
"loss": 0.2485,
"step": 1340
},
{
"epoch": 11.0,
"eval_accuracy": 0.9105858170606372,
"eval_loss": 0.30900049209594727,
"eval_runtime": 12.1706,
"eval_samples_per_second": 79.947,
"eval_steps_per_second": 5.012,
"step": 1342
},
{
"epoch": 11.02,
"learning_rate": 4.075342465753425e-05,
"loss": 0.3609,
"step": 1345
},
{
"epoch": 11.07,
"learning_rate": 3.904109589041096e-05,
"loss": 0.2121,
"step": 1350
},
{
"epoch": 11.11,
"learning_rate": 3.7328767123287675e-05,
"loss": 0.3091,
"step": 1355
},
{
"epoch": 11.15,
"learning_rate": 3.561643835616438e-05,
"loss": 0.317,
"step": 1360
},
{
"epoch": 11.19,
"learning_rate": 3.39041095890411e-05,
"loss": 0.4141,
"step": 1365
},
{
"epoch": 11.23,
"learning_rate": 3.219178082191781e-05,
"loss": 0.3378,
"step": 1370
},
{
"epoch": 11.27,
"learning_rate": 3.047945205479452e-05,
"loss": 0.3362,
"step": 1375
},
{
"epoch": 11.31,
"learning_rate": 2.8767123287671234e-05,
"loss": 0.2754,
"step": 1380
},
{
"epoch": 11.35,
"learning_rate": 2.7054794520547947e-05,
"loss": 0.3086,
"step": 1385
},
{
"epoch": 11.39,
"learning_rate": 2.534246575342466e-05,
"loss": 0.2916,
"step": 1390
},
{
"epoch": 11.43,
"learning_rate": 2.363013698630137e-05,
"loss": 0.272,
"step": 1395
},
{
"epoch": 11.48,
"learning_rate": 2.1917808219178083e-05,
"loss": 0.2979,
"step": 1400
},
{
"epoch": 11.52,
"learning_rate": 2.0205479452054797e-05,
"loss": 0.2144,
"step": 1405
},
{
"epoch": 11.56,
"learning_rate": 1.8493150684931506e-05,
"loss": 0.1952,
"step": 1410
},
{
"epoch": 11.6,
"learning_rate": 1.678082191780822e-05,
"loss": 0.1699,
"step": 1415
},
{
"epoch": 11.64,
"learning_rate": 1.5068493150684931e-05,
"loss": 0.2657,
"step": 1420
},
{
"epoch": 11.68,
"learning_rate": 1.3356164383561646e-05,
"loss": 0.1957,
"step": 1425
},
{
"epoch": 11.72,
"learning_rate": 1.1643835616438355e-05,
"loss": 0.2203,
"step": 1430
},
{
"epoch": 11.76,
"learning_rate": 9.931506849315069e-06,
"loss": 0.1756,
"step": 1435
},
{
"epoch": 11.8,
"learning_rate": 8.21917808219178e-06,
"loss": 0.2771,
"step": 1440
},
{
"epoch": 11.84,
"learning_rate": 6.506849315068493e-06,
"loss": 0.2101,
"step": 1445
},
{
"epoch": 11.89,
"learning_rate": 4.7945205479452054e-06,
"loss": 0.1185,
"step": 1450
},
{
"epoch": 11.93,
"learning_rate": 3.0821917808219177e-06,
"loss": 0.2966,
"step": 1455
},
{
"epoch": 11.97,
"learning_rate": 1.3698630136986302e-06,
"loss": 0.1759,
"step": 1460
},
{
"epoch": 12.0,
"eval_accuracy": 0.986639260020555,
"eval_loss": 0.09759720414876938,
"eval_runtime": 12.1496,
"eval_samples_per_second": 80.085,
"eval_steps_per_second": 5.021,
"step": 1464
},
{
"epoch": 12.0,
"step": 1464,
"total_flos": 8.283619840147513e+18,
"train_loss": 0.8713844098517152,
"train_runtime": 5015.9628,
"train_samples_per_second": 9.309,
"train_steps_per_second": 0.292
}
],
"max_steps": 1464,
"num_train_epochs": 12,
"total_flos": 8.283619840147513e+18,
"trial_name": null,
"trial_params": null
}