holylovenia's picture
Add model
de9fe44
{
"best_metric": 0.029699677601456642,
"best_model_checkpoint": "./save/eng-zho_elderly_facebook/wav2vec2-large-xlsr-53/checkpoint-2496",
"epoch": 22.0,
"global_step": 4576,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05,
"learning_rate": 4.997596153846154e-05,
"loss": 0.0533,
"step": 10
},
{
"epoch": 0.1,
"learning_rate": 4.995192307692308e-05,
"loss": 0.0523,
"step": 20
},
{
"epoch": 0.14,
"learning_rate": 4.992788461538462e-05,
"loss": 0.0517,
"step": 30
},
{
"epoch": 0.19,
"learning_rate": 4.9906250000000004e-05,
"loss": 0.0497,
"step": 40
},
{
"epoch": 0.24,
"learning_rate": 4.988221153846154e-05,
"loss": 0.0471,
"step": 50
},
{
"epoch": 0.29,
"learning_rate": 4.985817307692308e-05,
"loss": 0.0442,
"step": 60
},
{
"epoch": 0.34,
"learning_rate": 4.983413461538462e-05,
"loss": 0.0416,
"step": 70
},
{
"epoch": 0.38,
"learning_rate": 4.9810096153846156e-05,
"loss": 0.0393,
"step": 80
},
{
"epoch": 0.43,
"learning_rate": 4.97860576923077e-05,
"loss": 0.0424,
"step": 90
},
{
"epoch": 0.48,
"learning_rate": 4.9762019230769235e-05,
"loss": 0.0414,
"step": 100
},
{
"epoch": 0.53,
"learning_rate": 4.973798076923077e-05,
"loss": 0.0394,
"step": 110
},
{
"epoch": 0.58,
"learning_rate": 4.9713942307692314e-05,
"loss": 0.0389,
"step": 120
},
{
"epoch": 0.62,
"learning_rate": 4.968990384615385e-05,
"loss": 0.0427,
"step": 130
},
{
"epoch": 0.67,
"learning_rate": 4.966586538461539e-05,
"loss": 0.0421,
"step": 140
},
{
"epoch": 0.72,
"learning_rate": 4.964182692307692e-05,
"loss": 0.0402,
"step": 150
},
{
"epoch": 0.77,
"learning_rate": 4.961778846153846e-05,
"loss": 0.0412,
"step": 160
},
{
"epoch": 0.82,
"learning_rate": 4.959375e-05,
"loss": 0.0404,
"step": 170
},
{
"epoch": 0.87,
"learning_rate": 4.956971153846154e-05,
"loss": 0.0408,
"step": 180
},
{
"epoch": 0.91,
"learning_rate": 4.9545673076923075e-05,
"loss": 0.0391,
"step": 190
},
{
"epoch": 0.96,
"learning_rate": 4.952163461538462e-05,
"loss": 0.0378,
"step": 200
},
{
"epoch": 1.0,
"eval_loss": 0.03970440477132797,
"eval_runtime": 37.2853,
"eval_samples_per_second": 17.406,
"eval_steps_per_second": 2.199,
"step": 208
},
{
"epoch": 1.01,
"learning_rate": 4.9497596153846155e-05,
"loss": 0.0407,
"step": 210
},
{
"epoch": 1.06,
"learning_rate": 4.947355769230769e-05,
"loss": 0.0404,
"step": 220
},
{
"epoch": 1.11,
"learning_rate": 4.9449519230769234e-05,
"loss": 0.0406,
"step": 230
},
{
"epoch": 1.15,
"learning_rate": 4.942548076923077e-05,
"loss": 0.0377,
"step": 240
},
{
"epoch": 1.2,
"learning_rate": 4.940144230769231e-05,
"loss": 0.0401,
"step": 250
},
{
"epoch": 1.25,
"learning_rate": 4.937740384615385e-05,
"loss": 0.0391,
"step": 260
},
{
"epoch": 1.3,
"learning_rate": 4.9353365384615386e-05,
"loss": 0.042,
"step": 270
},
{
"epoch": 1.35,
"learning_rate": 4.932932692307692e-05,
"loss": 0.0371,
"step": 280
},
{
"epoch": 1.39,
"learning_rate": 4.9305288461538466e-05,
"loss": 0.0405,
"step": 290
},
{
"epoch": 1.44,
"learning_rate": 4.928125e-05,
"loss": 0.038,
"step": 300
},
{
"epoch": 1.49,
"learning_rate": 4.925721153846154e-05,
"loss": 0.0384,
"step": 310
},
{
"epoch": 1.54,
"learning_rate": 4.923317307692308e-05,
"loss": 0.0398,
"step": 320
},
{
"epoch": 1.59,
"learning_rate": 4.920913461538462e-05,
"loss": 0.037,
"step": 330
},
{
"epoch": 1.63,
"learning_rate": 4.9185096153846154e-05,
"loss": 0.0389,
"step": 340
},
{
"epoch": 1.68,
"learning_rate": 4.91610576923077e-05,
"loss": 0.037,
"step": 350
},
{
"epoch": 1.73,
"learning_rate": 4.913701923076923e-05,
"loss": 0.0375,
"step": 360
},
{
"epoch": 1.78,
"learning_rate": 4.911298076923077e-05,
"loss": 0.0362,
"step": 370
},
{
"epoch": 1.83,
"learning_rate": 4.908894230769231e-05,
"loss": 0.0405,
"step": 380
},
{
"epoch": 1.88,
"learning_rate": 4.906490384615385e-05,
"loss": 0.0374,
"step": 390
},
{
"epoch": 1.92,
"learning_rate": 4.9040865384615385e-05,
"loss": 0.0374,
"step": 400
},
{
"epoch": 1.97,
"learning_rate": 4.901682692307693e-05,
"loss": 0.036,
"step": 410
},
{
"epoch": 2.0,
"eval_loss": 0.037033889442682266,
"eval_runtime": 36.6058,
"eval_samples_per_second": 17.729,
"eval_steps_per_second": 2.24,
"step": 416
},
{
"epoch": 2.02,
"learning_rate": 4.8992788461538465e-05,
"loss": 0.0362,
"step": 420
},
{
"epoch": 2.07,
"learning_rate": 4.896875e-05,
"loss": 0.0391,
"step": 430
},
{
"epoch": 2.12,
"learning_rate": 4.894471153846154e-05,
"loss": 0.0361,
"step": 440
},
{
"epoch": 2.16,
"learning_rate": 4.892067307692308e-05,
"loss": 0.0384,
"step": 450
},
{
"epoch": 2.21,
"learning_rate": 4.889663461538462e-05,
"loss": 0.0387,
"step": 460
},
{
"epoch": 2.26,
"learning_rate": 4.887259615384615e-05,
"loss": 0.0333,
"step": 470
},
{
"epoch": 2.31,
"learning_rate": 4.8848557692307696e-05,
"loss": 0.0371,
"step": 480
},
{
"epoch": 2.36,
"learning_rate": 4.882451923076923e-05,
"loss": 0.0351,
"step": 490
},
{
"epoch": 2.4,
"learning_rate": 4.880048076923077e-05,
"loss": 0.0356,
"step": 500
},
{
"epoch": 2.45,
"learning_rate": 4.877644230769231e-05,
"loss": 0.0355,
"step": 510
},
{
"epoch": 2.5,
"learning_rate": 4.875240384615385e-05,
"loss": 0.0369,
"step": 520
},
{
"epoch": 2.55,
"learning_rate": 4.8728365384615385e-05,
"loss": 0.0349,
"step": 530
},
{
"epoch": 2.6,
"learning_rate": 4.870432692307693e-05,
"loss": 0.0358,
"step": 540
},
{
"epoch": 2.64,
"learning_rate": 4.8680288461538464e-05,
"loss": 0.0344,
"step": 550
},
{
"epoch": 2.69,
"learning_rate": 4.865625e-05,
"loss": 0.0341,
"step": 560
},
{
"epoch": 2.74,
"learning_rate": 4.863221153846154e-05,
"loss": 0.0367,
"step": 570
},
{
"epoch": 2.79,
"learning_rate": 4.860817307692308e-05,
"loss": 0.0345,
"step": 580
},
{
"epoch": 2.84,
"learning_rate": 4.8584134615384616e-05,
"loss": 0.0365,
"step": 590
},
{
"epoch": 2.88,
"learning_rate": 4.856009615384616e-05,
"loss": 0.0334,
"step": 600
},
{
"epoch": 2.93,
"learning_rate": 4.8536057692307695e-05,
"loss": 0.0356,
"step": 610
},
{
"epoch": 2.98,
"learning_rate": 4.851201923076923e-05,
"loss": 0.0322,
"step": 620
},
{
"epoch": 3.0,
"eval_loss": 0.034713711589574814,
"eval_runtime": 36.4637,
"eval_samples_per_second": 17.799,
"eval_steps_per_second": 2.249,
"step": 624
},
{
"epoch": 3.03,
"learning_rate": 4.8487980769230775e-05,
"loss": 0.0354,
"step": 630
},
{
"epoch": 3.08,
"learning_rate": 4.846394230769231e-05,
"loss": 0.0315,
"step": 640
},
{
"epoch": 3.12,
"learning_rate": 4.843990384615385e-05,
"loss": 0.036,
"step": 650
},
{
"epoch": 3.17,
"learning_rate": 4.841586538461539e-05,
"loss": 0.0358,
"step": 660
},
{
"epoch": 3.22,
"learning_rate": 4.839182692307693e-05,
"loss": 0.0336,
"step": 670
},
{
"epoch": 3.27,
"learning_rate": 4.836778846153846e-05,
"loss": 0.0349,
"step": 680
},
{
"epoch": 3.32,
"learning_rate": 4.8343750000000006e-05,
"loss": 0.0342,
"step": 690
},
{
"epoch": 3.37,
"learning_rate": 4.831971153846154e-05,
"loss": 0.0404,
"step": 700
},
{
"epoch": 3.41,
"learning_rate": 4.829567307692308e-05,
"loss": 0.0317,
"step": 710
},
{
"epoch": 3.46,
"learning_rate": 4.827163461538462e-05,
"loss": 0.0362,
"step": 720
},
{
"epoch": 3.51,
"learning_rate": 4.824759615384616e-05,
"loss": 0.03,
"step": 730
},
{
"epoch": 3.56,
"learning_rate": 4.8223557692307695e-05,
"loss": 0.0311,
"step": 740
},
{
"epoch": 3.61,
"learning_rate": 4.819951923076924e-05,
"loss": 0.033,
"step": 750
},
{
"epoch": 3.65,
"learning_rate": 4.8175480769230774e-05,
"loss": 0.0372,
"step": 760
},
{
"epoch": 3.7,
"learning_rate": 4.815144230769231e-05,
"loss": 0.0349,
"step": 770
},
{
"epoch": 3.75,
"learning_rate": 4.812740384615385e-05,
"loss": 0.0329,
"step": 780
},
{
"epoch": 3.8,
"learning_rate": 4.810336538461538e-05,
"loss": 0.0324,
"step": 790
},
{
"epoch": 3.85,
"learning_rate": 4.8079326923076926e-05,
"loss": 0.0313,
"step": 800
},
{
"epoch": 3.89,
"learning_rate": 4.805528846153846e-05,
"loss": 0.0321,
"step": 810
},
{
"epoch": 3.94,
"learning_rate": 4.803125e-05,
"loss": 0.0365,
"step": 820
},
{
"epoch": 3.99,
"learning_rate": 4.800721153846154e-05,
"loss": 0.0299,
"step": 830
},
{
"epoch": 4.0,
"eval_loss": 0.033535003662109375,
"eval_runtime": 36.6636,
"eval_samples_per_second": 17.701,
"eval_steps_per_second": 2.237,
"step": 832
},
{
"epoch": 4.04,
"learning_rate": 4.798317307692308e-05,
"loss": 0.0355,
"step": 840
},
{
"epoch": 4.09,
"learning_rate": 4.7959134615384614e-05,
"loss": 0.0303,
"step": 850
},
{
"epoch": 4.13,
"learning_rate": 4.793509615384616e-05,
"loss": 0.0326,
"step": 860
},
{
"epoch": 4.18,
"learning_rate": 4.7911057692307694e-05,
"loss": 0.0308,
"step": 870
},
{
"epoch": 4.23,
"learning_rate": 4.788701923076923e-05,
"loss": 0.0325,
"step": 880
},
{
"epoch": 4.28,
"learning_rate": 4.7862980769230766e-05,
"loss": 0.0357,
"step": 890
},
{
"epoch": 4.33,
"learning_rate": 4.783894230769231e-05,
"loss": 0.0337,
"step": 900
},
{
"epoch": 4.38,
"learning_rate": 4.7814903846153846e-05,
"loss": 0.0343,
"step": 910
},
{
"epoch": 4.42,
"learning_rate": 4.779086538461538e-05,
"loss": 0.0287,
"step": 920
},
{
"epoch": 4.47,
"learning_rate": 4.7766826923076925e-05,
"loss": 0.0294,
"step": 930
},
{
"epoch": 4.52,
"learning_rate": 4.774278846153846e-05,
"loss": 0.0289,
"step": 940
},
{
"epoch": 4.57,
"learning_rate": 4.771875e-05,
"loss": 0.03,
"step": 950
},
{
"epoch": 4.62,
"learning_rate": 4.769471153846154e-05,
"loss": 0.0337,
"step": 960
},
{
"epoch": 4.66,
"learning_rate": 4.767067307692308e-05,
"loss": 0.0335,
"step": 970
},
{
"epoch": 4.71,
"learning_rate": 4.7646634615384613e-05,
"loss": 0.028,
"step": 980
},
{
"epoch": 4.76,
"learning_rate": 4.7622596153846157e-05,
"loss": 0.0304,
"step": 990
},
{
"epoch": 4.81,
"learning_rate": 4.759855769230769e-05,
"loss": 0.0324,
"step": 1000
},
{
"epoch": 4.86,
"learning_rate": 4.757451923076923e-05,
"loss": 0.0282,
"step": 1010
},
{
"epoch": 4.9,
"learning_rate": 4.755048076923077e-05,
"loss": 0.0306,
"step": 1020
},
{
"epoch": 4.95,
"learning_rate": 4.752644230769231e-05,
"loss": 0.0304,
"step": 1030
},
{
"epoch": 5.0,
"learning_rate": 4.7502403846153845e-05,
"loss": 0.0323,
"step": 1040
},
{
"epoch": 5.0,
"eval_loss": 0.032066214829683304,
"eval_runtime": 36.0368,
"eval_samples_per_second": 18.009,
"eval_steps_per_second": 2.275,
"step": 1040
},
{
"epoch": 5.05,
"learning_rate": 4.747836538461539e-05,
"loss": 0.0298,
"step": 1050
},
{
"epoch": 5.1,
"learning_rate": 4.7454326923076924e-05,
"loss": 0.0317,
"step": 1060
},
{
"epoch": 5.14,
"learning_rate": 4.743028846153846e-05,
"loss": 0.0309,
"step": 1070
},
{
"epoch": 5.19,
"learning_rate": 4.7406250000000004e-05,
"loss": 0.0315,
"step": 1080
},
{
"epoch": 5.24,
"learning_rate": 4.738221153846154e-05,
"loss": 0.0279,
"step": 1090
},
{
"epoch": 5.29,
"learning_rate": 4.7358173076923076e-05,
"loss": 0.0287,
"step": 1100
},
{
"epoch": 5.34,
"learning_rate": 4.733413461538462e-05,
"loss": 0.0286,
"step": 1110
},
{
"epoch": 5.38,
"learning_rate": 4.7310096153846156e-05,
"loss": 0.0301,
"step": 1120
},
{
"epoch": 5.43,
"learning_rate": 4.728605769230769e-05,
"loss": 0.0345,
"step": 1130
},
{
"epoch": 5.48,
"learning_rate": 4.7262019230769235e-05,
"loss": 0.03,
"step": 1140
},
{
"epoch": 5.53,
"learning_rate": 4.723798076923077e-05,
"loss": 0.0263,
"step": 1150
},
{
"epoch": 5.58,
"learning_rate": 4.721394230769231e-05,
"loss": 0.0278,
"step": 1160
},
{
"epoch": 5.62,
"learning_rate": 4.718990384615385e-05,
"loss": 0.0281,
"step": 1170
},
{
"epoch": 5.67,
"learning_rate": 4.716586538461539e-05,
"loss": 0.0304,
"step": 1180
},
{
"epoch": 5.72,
"learning_rate": 4.7141826923076924e-05,
"loss": 0.0269,
"step": 1190
},
{
"epoch": 5.77,
"learning_rate": 4.7117788461538467e-05,
"loss": 0.0301,
"step": 1200
},
{
"epoch": 5.82,
"learning_rate": 4.709375e-05,
"loss": 0.0284,
"step": 1210
},
{
"epoch": 5.87,
"learning_rate": 4.706971153846154e-05,
"loss": 0.0292,
"step": 1220
},
{
"epoch": 5.91,
"learning_rate": 4.704567307692308e-05,
"loss": 0.0304,
"step": 1230
},
{
"epoch": 5.96,
"learning_rate": 4.702163461538462e-05,
"loss": 0.0328,
"step": 1240
},
{
"epoch": 6.0,
"eval_loss": 0.030981002375483513,
"eval_runtime": 36.8216,
"eval_samples_per_second": 17.626,
"eval_steps_per_second": 2.227,
"step": 1248
},
{
"epoch": 6.01,
"learning_rate": 4.6997596153846155e-05,
"loss": 0.0265,
"step": 1250
},
{
"epoch": 6.06,
"learning_rate": 4.69735576923077e-05,
"loss": 0.029,
"step": 1260
},
{
"epoch": 6.11,
"learning_rate": 4.6949519230769234e-05,
"loss": 0.0273,
"step": 1270
},
{
"epoch": 6.15,
"learning_rate": 4.692548076923077e-05,
"loss": 0.0275,
"step": 1280
},
{
"epoch": 6.2,
"learning_rate": 4.6901442307692314e-05,
"loss": 0.0283,
"step": 1290
},
{
"epoch": 6.25,
"learning_rate": 4.687740384615385e-05,
"loss": 0.025,
"step": 1300
},
{
"epoch": 6.3,
"learning_rate": 4.6853365384615386e-05,
"loss": 0.023,
"step": 1310
},
{
"epoch": 6.35,
"learning_rate": 4.682932692307693e-05,
"loss": 0.0275,
"step": 1320
},
{
"epoch": 6.39,
"learning_rate": 4.6805288461538466e-05,
"loss": 0.0286,
"step": 1330
},
{
"epoch": 6.44,
"learning_rate": 4.678125e-05,
"loss": 0.0267,
"step": 1340
},
{
"epoch": 6.49,
"learning_rate": 4.6757211538461545e-05,
"loss": 0.031,
"step": 1350
},
{
"epoch": 6.54,
"learning_rate": 4.673317307692308e-05,
"loss": 0.0267,
"step": 1360
},
{
"epoch": 6.59,
"learning_rate": 4.670913461538462e-05,
"loss": 0.0283,
"step": 1370
},
{
"epoch": 6.63,
"learning_rate": 4.668509615384616e-05,
"loss": 0.0266,
"step": 1380
},
{
"epoch": 6.68,
"learning_rate": 4.66610576923077e-05,
"loss": 0.0277,
"step": 1390
},
{
"epoch": 6.73,
"learning_rate": 4.6637019230769234e-05,
"loss": 0.0298,
"step": 1400
},
{
"epoch": 6.78,
"learning_rate": 4.6612980769230777e-05,
"loss": 0.0294,
"step": 1410
},
{
"epoch": 6.83,
"learning_rate": 4.658894230769231e-05,
"loss": 0.0253,
"step": 1420
},
{
"epoch": 6.88,
"learning_rate": 4.656490384615385e-05,
"loss": 0.0264,
"step": 1430
},
{
"epoch": 6.92,
"learning_rate": 4.6540865384615386e-05,
"loss": 0.0288,
"step": 1440
},
{
"epoch": 6.97,
"learning_rate": 4.651682692307692e-05,
"loss": 0.024,
"step": 1450
},
{
"epoch": 7.0,
"eval_loss": 0.031017007306218147,
"eval_runtime": 37.113,
"eval_samples_per_second": 17.487,
"eval_steps_per_second": 2.209,
"step": 1456
},
{
"epoch": 7.02,
"learning_rate": 4.649278846153846e-05,
"loss": 0.0238,
"step": 1460
},
{
"epoch": 7.07,
"learning_rate": 4.646875e-05,
"loss": 0.0284,
"step": 1470
},
{
"epoch": 7.12,
"learning_rate": 4.644471153846154e-05,
"loss": 0.0248,
"step": 1480
},
{
"epoch": 7.16,
"learning_rate": 4.6420673076923074e-05,
"loss": 0.0234,
"step": 1490
},
{
"epoch": 7.21,
"learning_rate": 4.639663461538462e-05,
"loss": 0.0271,
"step": 1500
},
{
"epoch": 7.26,
"learning_rate": 4.637259615384615e-05,
"loss": 0.0267,
"step": 1510
},
{
"epoch": 7.31,
"learning_rate": 4.634855769230769e-05,
"loss": 0.0252,
"step": 1520
},
{
"epoch": 7.36,
"learning_rate": 4.632451923076923e-05,
"loss": 0.0257,
"step": 1530
},
{
"epoch": 7.4,
"learning_rate": 4.630048076923077e-05,
"loss": 0.0224,
"step": 1540
},
{
"epoch": 7.45,
"learning_rate": 4.6276442307692305e-05,
"loss": 0.0277,
"step": 1550
},
{
"epoch": 7.5,
"learning_rate": 4.625240384615385e-05,
"loss": 0.0297,
"step": 1560
},
{
"epoch": 7.55,
"learning_rate": 4.6228365384615385e-05,
"loss": 0.0254,
"step": 1570
},
{
"epoch": 7.6,
"learning_rate": 4.620432692307692e-05,
"loss": 0.0238,
"step": 1580
},
{
"epoch": 7.64,
"learning_rate": 4.6180288461538464e-05,
"loss": 0.0243,
"step": 1590
},
{
"epoch": 7.69,
"learning_rate": 4.615625e-05,
"loss": 0.0262,
"step": 1600
},
{
"epoch": 7.74,
"learning_rate": 4.613221153846154e-05,
"loss": 0.0308,
"step": 1610
},
{
"epoch": 7.79,
"learning_rate": 4.610817307692308e-05,
"loss": 0.022,
"step": 1620
},
{
"epoch": 7.84,
"learning_rate": 4.6084134615384616e-05,
"loss": 0.0236,
"step": 1630
},
{
"epoch": 7.88,
"learning_rate": 4.606009615384615e-05,
"loss": 0.0251,
"step": 1640
},
{
"epoch": 7.93,
"learning_rate": 4.6036057692307696e-05,
"loss": 0.0261,
"step": 1650
},
{
"epoch": 7.98,
"learning_rate": 4.601201923076923e-05,
"loss": 0.024,
"step": 1660
},
{
"epoch": 8.0,
"eval_loss": 0.030306896194815636,
"eval_runtime": 37.1939,
"eval_samples_per_second": 17.449,
"eval_steps_per_second": 2.205,
"step": 1664
},
{
"epoch": 8.03,
"learning_rate": 4.598798076923077e-05,
"loss": 0.0273,
"step": 1670
},
{
"epoch": 8.08,
"learning_rate": 4.596394230769231e-05,
"loss": 0.0256,
"step": 1680
},
{
"epoch": 8.12,
"learning_rate": 4.593990384615385e-05,
"loss": 0.0246,
"step": 1690
},
{
"epoch": 8.17,
"learning_rate": 4.5915865384615384e-05,
"loss": 0.0216,
"step": 1700
},
{
"epoch": 8.22,
"learning_rate": 4.589182692307693e-05,
"loss": 0.0237,
"step": 1710
},
{
"epoch": 8.27,
"learning_rate": 4.586778846153846e-05,
"loss": 0.0205,
"step": 1720
},
{
"epoch": 8.32,
"learning_rate": 4.584375e-05,
"loss": 0.0221,
"step": 1730
},
{
"epoch": 8.37,
"learning_rate": 4.581971153846154e-05,
"loss": 0.0282,
"step": 1740
},
{
"epoch": 8.41,
"learning_rate": 4.579567307692308e-05,
"loss": 0.0283,
"step": 1750
},
{
"epoch": 8.46,
"learning_rate": 4.5771634615384615e-05,
"loss": 0.0214,
"step": 1760
},
{
"epoch": 8.51,
"learning_rate": 4.574759615384616e-05,
"loss": 0.0267,
"step": 1770
},
{
"epoch": 8.56,
"learning_rate": 4.5723557692307695e-05,
"loss": 0.02,
"step": 1780
},
{
"epoch": 8.61,
"learning_rate": 4.569951923076923e-05,
"loss": 0.0234,
"step": 1790
},
{
"epoch": 8.65,
"learning_rate": 4.5675480769230774e-05,
"loss": 0.0269,
"step": 1800
},
{
"epoch": 8.7,
"learning_rate": 4.565144230769231e-05,
"loss": 0.0255,
"step": 1810
},
{
"epoch": 8.75,
"learning_rate": 4.562740384615385e-05,
"loss": 0.0256,
"step": 1820
},
{
"epoch": 8.8,
"learning_rate": 4.560336538461539e-05,
"loss": 0.0227,
"step": 1830
},
{
"epoch": 8.85,
"learning_rate": 4.5579326923076926e-05,
"loss": 0.0233,
"step": 1840
},
{
"epoch": 8.89,
"learning_rate": 4.555528846153846e-05,
"loss": 0.0235,
"step": 1850
},
{
"epoch": 8.94,
"learning_rate": 4.5531250000000006e-05,
"loss": 0.029,
"step": 1860
},
{
"epoch": 8.99,
"learning_rate": 4.550721153846154e-05,
"loss": 0.0252,
"step": 1870
},
{
"epoch": 9.0,
"eval_loss": 0.030633406713604927,
"eval_runtime": 37.8394,
"eval_samples_per_second": 17.151,
"eval_steps_per_second": 2.167,
"step": 1872
},
{
"epoch": 9.04,
"learning_rate": 4.548317307692308e-05,
"loss": 0.0241,
"step": 1880
},
{
"epoch": 9.09,
"learning_rate": 4.545913461538462e-05,
"loss": 0.0207,
"step": 1890
},
{
"epoch": 9.13,
"learning_rate": 4.543509615384616e-05,
"loss": 0.0219,
"step": 1900
},
{
"epoch": 9.18,
"learning_rate": 4.5411057692307694e-05,
"loss": 0.0226,
"step": 1910
},
{
"epoch": 9.23,
"learning_rate": 4.538701923076924e-05,
"loss": 0.0205,
"step": 1920
},
{
"epoch": 9.28,
"learning_rate": 4.536298076923077e-05,
"loss": 0.0209,
"step": 1930
},
{
"epoch": 9.33,
"learning_rate": 4.533894230769231e-05,
"loss": 0.0232,
"step": 1940
},
{
"epoch": 9.38,
"learning_rate": 4.531490384615385e-05,
"loss": 0.0241,
"step": 1950
},
{
"epoch": 9.42,
"learning_rate": 4.529086538461539e-05,
"loss": 0.022,
"step": 1960
},
{
"epoch": 9.47,
"learning_rate": 4.5266826923076925e-05,
"loss": 0.0231,
"step": 1970
},
{
"epoch": 9.52,
"learning_rate": 4.524278846153847e-05,
"loss": 0.0242,
"step": 1980
},
{
"epoch": 9.57,
"learning_rate": 4.5218750000000005e-05,
"loss": 0.0247,
"step": 1990
},
{
"epoch": 9.62,
"learning_rate": 4.519471153846154e-05,
"loss": 0.0225,
"step": 2000
},
{
"epoch": 9.66,
"learning_rate": 4.5170673076923084e-05,
"loss": 0.032,
"step": 2010
},
{
"epoch": 9.71,
"learning_rate": 4.514663461538462e-05,
"loss": 0.022,
"step": 2020
},
{
"epoch": 9.76,
"learning_rate": 4.512259615384616e-05,
"loss": 0.0184,
"step": 2030
},
{
"epoch": 9.81,
"learning_rate": 4.50985576923077e-05,
"loss": 0.0235,
"step": 2040
},
{
"epoch": 9.86,
"learning_rate": 4.5074519230769236e-05,
"loss": 0.0215,
"step": 2050
},
{
"epoch": 9.9,
"learning_rate": 4.505048076923077e-05,
"loss": 0.0241,
"step": 2060
},
{
"epoch": 9.95,
"learning_rate": 4.5026442307692316e-05,
"loss": 0.0193,
"step": 2070
},
{
"epoch": 10.0,
"learning_rate": 4.500240384615385e-05,
"loss": 0.0203,
"step": 2080
},
{
"epoch": 10.0,
"eval_loss": 0.030458811670541763,
"eval_runtime": 36.2671,
"eval_samples_per_second": 17.895,
"eval_steps_per_second": 2.261,
"step": 2080
},
{
"epoch": 10.05,
"learning_rate": 4.497836538461538e-05,
"loss": 0.0231,
"step": 2090
},
{
"epoch": 10.1,
"learning_rate": 4.4954326923076925e-05,
"loss": 0.0237,
"step": 2100
},
{
"epoch": 10.14,
"learning_rate": 4.493028846153846e-05,
"loss": 0.0193,
"step": 2110
},
{
"epoch": 10.19,
"learning_rate": 4.490625e-05,
"loss": 0.0202,
"step": 2120
},
{
"epoch": 10.24,
"learning_rate": 4.488221153846154e-05,
"loss": 0.0246,
"step": 2130
},
{
"epoch": 10.29,
"learning_rate": 4.4858173076923077e-05,
"loss": 0.0245,
"step": 2140
},
{
"epoch": 10.34,
"learning_rate": 4.483413461538461e-05,
"loss": 0.02,
"step": 2150
},
{
"epoch": 10.38,
"learning_rate": 4.4810096153846156e-05,
"loss": 0.0224,
"step": 2160
},
{
"epoch": 10.43,
"learning_rate": 4.478605769230769e-05,
"loss": 0.0209,
"step": 2170
},
{
"epoch": 10.48,
"learning_rate": 4.476201923076923e-05,
"loss": 0.02,
"step": 2180
},
{
"epoch": 10.53,
"learning_rate": 4.473798076923077e-05,
"loss": 0.0233,
"step": 2190
},
{
"epoch": 10.58,
"learning_rate": 4.471634615384616e-05,
"loss": 0.0222,
"step": 2200
},
{
"epoch": 10.62,
"learning_rate": 4.4692307692307693e-05,
"loss": 0.0233,
"step": 2210
},
{
"epoch": 10.67,
"learning_rate": 4.466826923076923e-05,
"loss": 0.0216,
"step": 2220
},
{
"epoch": 10.72,
"learning_rate": 4.464423076923077e-05,
"loss": 0.0239,
"step": 2230
},
{
"epoch": 10.77,
"learning_rate": 4.462019230769231e-05,
"loss": 0.019,
"step": 2240
},
{
"epoch": 10.82,
"learning_rate": 4.4596153846153845e-05,
"loss": 0.023,
"step": 2250
},
{
"epoch": 10.87,
"learning_rate": 4.457211538461539e-05,
"loss": 0.0218,
"step": 2260
},
{
"epoch": 10.91,
"learning_rate": 4.4548076923076925e-05,
"loss": 0.0223,
"step": 2270
},
{
"epoch": 10.96,
"learning_rate": 4.452403846153846e-05,
"loss": 0.0175,
"step": 2280
},
{
"epoch": 11.0,
"eval_loss": 0.030906477943062782,
"eval_runtime": 36.4911,
"eval_samples_per_second": 17.785,
"eval_steps_per_second": 2.247,
"step": 2288
},
{
"epoch": 11.01,
"learning_rate": 4.4500000000000004e-05,
"loss": 0.0194,
"step": 2290
},
{
"epoch": 11.06,
"learning_rate": 4.447596153846154e-05,
"loss": 0.0205,
"step": 2300
},
{
"epoch": 11.11,
"learning_rate": 4.445192307692308e-05,
"loss": 0.0234,
"step": 2310
},
{
"epoch": 11.15,
"learning_rate": 4.442788461538462e-05,
"loss": 0.0202,
"step": 2320
},
{
"epoch": 11.2,
"learning_rate": 4.4403846153846156e-05,
"loss": 0.0191,
"step": 2330
},
{
"epoch": 11.25,
"learning_rate": 4.437980769230769e-05,
"loss": 0.0176,
"step": 2340
},
{
"epoch": 11.3,
"learning_rate": 4.4355769230769236e-05,
"loss": 0.0175,
"step": 2350
},
{
"epoch": 11.35,
"learning_rate": 4.433173076923077e-05,
"loss": 0.0194,
"step": 2360
},
{
"epoch": 11.39,
"learning_rate": 4.430769230769231e-05,
"loss": 0.0177,
"step": 2370
},
{
"epoch": 11.44,
"learning_rate": 4.428365384615385e-05,
"loss": 0.0182,
"step": 2380
},
{
"epoch": 11.49,
"learning_rate": 4.425961538461539e-05,
"loss": 0.0225,
"step": 2390
},
{
"epoch": 11.54,
"learning_rate": 4.4235576923076924e-05,
"loss": 0.0189,
"step": 2400
},
{
"epoch": 11.59,
"learning_rate": 4.421153846153847e-05,
"loss": 0.0183,
"step": 2410
},
{
"epoch": 11.63,
"learning_rate": 4.4187500000000003e-05,
"loss": 0.0233,
"step": 2420
},
{
"epoch": 11.68,
"learning_rate": 4.416346153846154e-05,
"loss": 0.0208,
"step": 2430
},
{
"epoch": 11.73,
"learning_rate": 4.413942307692308e-05,
"loss": 0.0189,
"step": 2440
},
{
"epoch": 11.78,
"learning_rate": 4.411538461538462e-05,
"loss": 0.0201,
"step": 2450
},
{
"epoch": 11.83,
"learning_rate": 4.4091346153846155e-05,
"loss": 0.0223,
"step": 2460
},
{
"epoch": 11.88,
"learning_rate": 4.40673076923077e-05,
"loss": 0.0216,
"step": 2470
},
{
"epoch": 11.92,
"learning_rate": 4.4043269230769235e-05,
"loss": 0.0187,
"step": 2480
},
{
"epoch": 11.97,
"learning_rate": 4.4021634615384613e-05,
"loss": 0.0243,
"step": 2490
},
{
"epoch": 12.0,
"eval_loss": 0.029699677601456642,
"eval_runtime": 36.7497,
"eval_samples_per_second": 17.66,
"eval_steps_per_second": 2.231,
"step": 2496
},
{
"epoch": 12.02,
"learning_rate": 4.3997596153846157e-05,
"loss": 0.02,
"step": 2500
},
{
"epoch": 12.07,
"learning_rate": 4.397355769230769e-05,
"loss": 0.0165,
"step": 2510
},
{
"epoch": 12.12,
"learning_rate": 4.394951923076923e-05,
"loss": 0.0175,
"step": 2520
},
{
"epoch": 12.16,
"learning_rate": 4.392548076923077e-05,
"loss": 0.0178,
"step": 2530
},
{
"epoch": 12.21,
"learning_rate": 4.390144230769231e-05,
"loss": 0.0136,
"step": 2540
},
{
"epoch": 12.26,
"learning_rate": 4.3877403846153845e-05,
"loss": 0.0174,
"step": 2550
},
{
"epoch": 12.31,
"learning_rate": 4.385336538461539e-05,
"loss": 0.0209,
"step": 2560
},
{
"epoch": 12.36,
"learning_rate": 4.3829326923076924e-05,
"loss": 0.0191,
"step": 2570
},
{
"epoch": 12.4,
"learning_rate": 4.380528846153846e-05,
"loss": 0.0236,
"step": 2580
},
{
"epoch": 12.45,
"learning_rate": 4.3781250000000004e-05,
"loss": 0.0187,
"step": 2590
},
{
"epoch": 12.5,
"learning_rate": 4.375721153846154e-05,
"loss": 0.0182,
"step": 2600
},
{
"epoch": 12.55,
"learning_rate": 4.3733173076923076e-05,
"loss": 0.0142,
"step": 2610
},
{
"epoch": 12.6,
"learning_rate": 4.370913461538462e-05,
"loss": 0.0159,
"step": 2620
},
{
"epoch": 12.64,
"learning_rate": 4.3685096153846156e-05,
"loss": 0.0178,
"step": 2630
},
{
"epoch": 12.69,
"learning_rate": 4.366105769230769e-05,
"loss": 0.018,
"step": 2640
},
{
"epoch": 12.74,
"learning_rate": 4.3637019230769235e-05,
"loss": 0.0208,
"step": 2650
},
{
"epoch": 12.79,
"learning_rate": 4.361298076923077e-05,
"loss": 0.0181,
"step": 2660
},
{
"epoch": 12.84,
"learning_rate": 4.358894230769231e-05,
"loss": 0.0179,
"step": 2670
},
{
"epoch": 12.88,
"learning_rate": 4.356490384615385e-05,
"loss": 0.019,
"step": 2680
},
{
"epoch": 12.93,
"learning_rate": 4.354086538461539e-05,
"loss": 0.0203,
"step": 2690
},
{
"epoch": 12.98,
"learning_rate": 4.3516826923076923e-05,
"loss": 0.0197,
"step": 2700
},
{
"epoch": 13.0,
"eval_loss": 0.030876118689775467,
"eval_runtime": 36.3232,
"eval_samples_per_second": 17.867,
"eval_steps_per_second": 2.258,
"step": 2704
},
{
"epoch": 13.03,
"learning_rate": 4.3492788461538467e-05,
"loss": 0.0169,
"step": 2710
},
{
"epoch": 13.08,
"learning_rate": 4.346875e-05,
"loss": 0.0151,
"step": 2720
},
{
"epoch": 13.12,
"learning_rate": 4.344471153846154e-05,
"loss": 0.0147,
"step": 2730
},
{
"epoch": 13.17,
"learning_rate": 4.342067307692308e-05,
"loss": 0.0173,
"step": 2740
},
{
"epoch": 13.22,
"learning_rate": 4.339663461538462e-05,
"loss": 0.0183,
"step": 2750
},
{
"epoch": 13.27,
"learning_rate": 4.3372596153846155e-05,
"loss": 0.0158,
"step": 2760
},
{
"epoch": 13.32,
"learning_rate": 4.33485576923077e-05,
"loss": 0.0146,
"step": 2770
},
{
"epoch": 13.37,
"learning_rate": 4.3324519230769234e-05,
"loss": 0.0236,
"step": 2780
},
{
"epoch": 13.41,
"learning_rate": 4.330048076923077e-05,
"loss": 0.0154,
"step": 2790
},
{
"epoch": 13.46,
"learning_rate": 4.3276442307692314e-05,
"loss": 0.0166,
"step": 2800
},
{
"epoch": 13.51,
"learning_rate": 4.325240384615385e-05,
"loss": 0.018,
"step": 2810
},
{
"epoch": 13.56,
"learning_rate": 4.3228365384615386e-05,
"loss": 0.0166,
"step": 2820
},
{
"epoch": 13.61,
"learning_rate": 4.320432692307693e-05,
"loss": 0.0177,
"step": 2830
},
{
"epoch": 13.65,
"learning_rate": 4.3180288461538466e-05,
"loss": 0.0201,
"step": 2840
},
{
"epoch": 13.7,
"learning_rate": 4.315625e-05,
"loss": 0.02,
"step": 2850
},
{
"epoch": 13.75,
"learning_rate": 4.3132211538461545e-05,
"loss": 0.0243,
"step": 2860
},
{
"epoch": 13.8,
"learning_rate": 4.310817307692308e-05,
"loss": 0.0198,
"step": 2870
},
{
"epoch": 13.85,
"learning_rate": 4.308413461538462e-05,
"loss": 0.0182,
"step": 2880
},
{
"epoch": 13.89,
"learning_rate": 4.306009615384616e-05,
"loss": 0.0168,
"step": 2890
},
{
"epoch": 13.94,
"learning_rate": 4.30360576923077e-05,
"loss": 0.0161,
"step": 2900
},
{
"epoch": 13.99,
"learning_rate": 4.3012019230769233e-05,
"loss": 0.0154,
"step": 2910
},
{
"epoch": 14.0,
"eval_loss": 0.031427882611751556,
"eval_runtime": 36.8364,
"eval_samples_per_second": 17.618,
"eval_steps_per_second": 2.226,
"step": 2912
},
{
"epoch": 14.04,
"learning_rate": 4.2987980769230777e-05,
"loss": 0.0116,
"step": 2920
},
{
"epoch": 14.09,
"learning_rate": 4.296394230769231e-05,
"loss": 0.0126,
"step": 2930
},
{
"epoch": 14.13,
"learning_rate": 4.293990384615385e-05,
"loss": 0.0153,
"step": 2940
},
{
"epoch": 14.18,
"learning_rate": 4.291586538461539e-05,
"loss": 0.0166,
"step": 2950
},
{
"epoch": 14.23,
"learning_rate": 4.289182692307693e-05,
"loss": 0.0139,
"step": 2960
},
{
"epoch": 14.28,
"learning_rate": 4.2867788461538465e-05,
"loss": 0.0149,
"step": 2970
},
{
"epoch": 14.33,
"learning_rate": 4.284375000000001e-05,
"loss": 0.0205,
"step": 2980
},
{
"epoch": 14.38,
"learning_rate": 4.2819711538461544e-05,
"loss": 0.0156,
"step": 2990
},
{
"epoch": 14.42,
"learning_rate": 4.2795673076923074e-05,
"loss": 0.0158,
"step": 3000
},
{
"epoch": 14.47,
"learning_rate": 4.277163461538462e-05,
"loss": 0.0158,
"step": 3010
},
{
"epoch": 14.52,
"learning_rate": 4.274759615384615e-05,
"loss": 0.0208,
"step": 3020
},
{
"epoch": 14.57,
"learning_rate": 4.272355769230769e-05,
"loss": 0.0187,
"step": 3030
},
{
"epoch": 14.62,
"learning_rate": 4.269951923076923e-05,
"loss": 0.016,
"step": 3040
},
{
"epoch": 14.66,
"learning_rate": 4.267548076923077e-05,
"loss": 0.0163,
"step": 3050
},
{
"epoch": 14.71,
"learning_rate": 4.2651442307692305e-05,
"loss": 0.0152,
"step": 3060
},
{
"epoch": 14.76,
"learning_rate": 4.262740384615385e-05,
"loss": 0.016,
"step": 3070
},
{
"epoch": 14.81,
"learning_rate": 4.2603365384615385e-05,
"loss": 0.0124,
"step": 3080
},
{
"epoch": 14.86,
"learning_rate": 4.257932692307692e-05,
"loss": 0.0159,
"step": 3090
},
{
"epoch": 14.9,
"learning_rate": 4.2555288461538464e-05,
"loss": 0.0187,
"step": 3100
},
{
"epoch": 14.95,
"learning_rate": 4.253125e-05,
"loss": 0.0201,
"step": 3110
},
{
"epoch": 15.0,
"learning_rate": 4.250721153846154e-05,
"loss": 0.0123,
"step": 3120
},
{
"epoch": 15.0,
"eval_loss": 0.03161365166306496,
"eval_runtime": 36.679,
"eval_samples_per_second": 17.694,
"eval_steps_per_second": 2.236,
"step": 3120
},
{
"epoch": 15.05,
"learning_rate": 4.248317307692308e-05,
"loss": 0.0139,
"step": 3130
},
{
"epoch": 15.1,
"learning_rate": 4.2459134615384616e-05,
"loss": 0.0143,
"step": 3140
},
{
"epoch": 15.14,
"learning_rate": 4.243509615384615e-05,
"loss": 0.0156,
"step": 3150
},
{
"epoch": 15.19,
"learning_rate": 4.2411057692307696e-05,
"loss": 0.0134,
"step": 3160
},
{
"epoch": 15.24,
"learning_rate": 4.238701923076923e-05,
"loss": 0.0168,
"step": 3170
},
{
"epoch": 15.29,
"learning_rate": 4.236298076923077e-05,
"loss": 0.0119,
"step": 3180
},
{
"epoch": 15.34,
"learning_rate": 4.233894230769231e-05,
"loss": 0.0135,
"step": 3190
},
{
"epoch": 15.38,
"learning_rate": 4.231490384615385e-05,
"loss": 0.0159,
"step": 3200
},
{
"epoch": 15.43,
"learning_rate": 4.2290865384615384e-05,
"loss": 0.0122,
"step": 3210
},
{
"epoch": 15.48,
"learning_rate": 4.226682692307693e-05,
"loss": 0.0167,
"step": 3220
},
{
"epoch": 15.53,
"learning_rate": 4.224278846153846e-05,
"loss": 0.0142,
"step": 3230
},
{
"epoch": 15.58,
"learning_rate": 4.221875e-05,
"loss": 0.0159,
"step": 3240
},
{
"epoch": 15.62,
"learning_rate": 4.219471153846154e-05,
"loss": 0.0139,
"step": 3250
},
{
"epoch": 15.67,
"learning_rate": 4.217067307692308e-05,
"loss": 0.0111,
"step": 3260
},
{
"epoch": 15.72,
"learning_rate": 4.2146634615384615e-05,
"loss": 0.0116,
"step": 3270
},
{
"epoch": 15.77,
"learning_rate": 4.212259615384616e-05,
"loss": 0.0171,
"step": 3280
},
{
"epoch": 15.82,
"learning_rate": 4.2098557692307695e-05,
"loss": 0.0125,
"step": 3290
},
{
"epoch": 15.87,
"learning_rate": 4.207451923076923e-05,
"loss": 0.0187,
"step": 3300
},
{
"epoch": 15.91,
"learning_rate": 4.2050480769230774e-05,
"loss": 0.0113,
"step": 3310
},
{
"epoch": 15.96,
"learning_rate": 4.202644230769231e-05,
"loss": 0.0128,
"step": 3320
},
{
"epoch": 16.0,
"eval_loss": 0.03363659605383873,
"eval_runtime": 36.1996,
"eval_samples_per_second": 17.928,
"eval_steps_per_second": 2.265,
"step": 3328
},
{
"epoch": 16.01,
"learning_rate": 4.200240384615385e-05,
"loss": 0.0111,
"step": 3330
},
{
"epoch": 16.06,
"learning_rate": 4.197836538461539e-05,
"loss": 0.0129,
"step": 3340
},
{
"epoch": 16.11,
"learning_rate": 4.1954326923076926e-05,
"loss": 0.0149,
"step": 3350
},
{
"epoch": 16.15,
"learning_rate": 4.193028846153846e-05,
"loss": 0.0091,
"step": 3360
},
{
"epoch": 16.2,
"learning_rate": 4.1906250000000006e-05,
"loss": 0.0126,
"step": 3370
},
{
"epoch": 16.25,
"learning_rate": 4.188221153846154e-05,
"loss": 0.0103,
"step": 3380
},
{
"epoch": 16.3,
"learning_rate": 4.185817307692308e-05,
"loss": 0.0146,
"step": 3390
},
{
"epoch": 16.35,
"learning_rate": 4.183413461538462e-05,
"loss": 0.0134,
"step": 3400
},
{
"epoch": 16.39,
"learning_rate": 4.181009615384616e-05,
"loss": 0.0124,
"step": 3410
},
{
"epoch": 16.44,
"learning_rate": 4.1786057692307694e-05,
"loss": 0.0127,
"step": 3420
},
{
"epoch": 16.49,
"learning_rate": 4.176201923076924e-05,
"loss": 0.0165,
"step": 3430
},
{
"epoch": 16.54,
"learning_rate": 4.173798076923077e-05,
"loss": 0.0154,
"step": 3440
},
{
"epoch": 16.59,
"learning_rate": 4.171394230769231e-05,
"loss": 0.0126,
"step": 3450
},
{
"epoch": 16.63,
"learning_rate": 4.168990384615385e-05,
"loss": 0.013,
"step": 3460
},
{
"epoch": 16.68,
"learning_rate": 4.166586538461539e-05,
"loss": 0.0146,
"step": 3470
},
{
"epoch": 16.73,
"learning_rate": 4.1641826923076925e-05,
"loss": 0.0128,
"step": 3480
},
{
"epoch": 16.78,
"learning_rate": 4.161778846153846e-05,
"loss": 0.0119,
"step": 3490
},
{
"epoch": 16.83,
"learning_rate": 4.1593750000000005e-05,
"loss": 0.0114,
"step": 3500
},
{
"epoch": 16.88,
"learning_rate": 4.156971153846154e-05,
"loss": 0.0125,
"step": 3510
},
{
"epoch": 16.92,
"learning_rate": 4.154567307692308e-05,
"loss": 0.016,
"step": 3520
},
{
"epoch": 16.97,
"learning_rate": 4.152163461538462e-05,
"loss": 0.0136,
"step": 3530
},
{
"epoch": 17.0,
"eval_loss": 0.03450627624988556,
"eval_runtime": 36.2674,
"eval_samples_per_second": 17.895,
"eval_steps_per_second": 2.261,
"step": 3536
},
{
"epoch": 17.02,
"learning_rate": 4.149759615384616e-05,
"loss": 0.02,
"step": 3540
},
{
"epoch": 17.07,
"learning_rate": 4.147355769230769e-05,
"loss": 0.012,
"step": 3550
},
{
"epoch": 17.12,
"learning_rate": 4.1449519230769236e-05,
"loss": 0.0096,
"step": 3560
},
{
"epoch": 17.16,
"learning_rate": 4.142548076923077e-05,
"loss": 0.0141,
"step": 3570
},
{
"epoch": 17.21,
"learning_rate": 4.140144230769231e-05,
"loss": 0.0109,
"step": 3580
},
{
"epoch": 17.26,
"learning_rate": 4.137740384615385e-05,
"loss": 0.013,
"step": 3590
},
{
"epoch": 17.31,
"learning_rate": 4.135336538461539e-05,
"loss": 0.0121,
"step": 3600
},
{
"epoch": 17.36,
"learning_rate": 4.1329326923076925e-05,
"loss": 0.0144,
"step": 3610
},
{
"epoch": 17.4,
"learning_rate": 4.130528846153847e-05,
"loss": 0.0122,
"step": 3620
},
{
"epoch": 17.45,
"learning_rate": 4.1281250000000004e-05,
"loss": 0.0151,
"step": 3630
},
{
"epoch": 17.5,
"learning_rate": 4.125721153846154e-05,
"loss": 0.0109,
"step": 3640
},
{
"epoch": 17.55,
"learning_rate": 4.1233173076923077e-05,
"loss": 0.0117,
"step": 3650
},
{
"epoch": 17.6,
"learning_rate": 4.120913461538461e-05,
"loss": 0.0118,
"step": 3660
},
{
"epoch": 17.64,
"learning_rate": 4.1185096153846156e-05,
"loss": 0.0117,
"step": 3670
},
{
"epoch": 17.69,
"learning_rate": 4.116105769230769e-05,
"loss": 0.0156,
"step": 3680
},
{
"epoch": 17.74,
"learning_rate": 4.113701923076923e-05,
"loss": 0.0168,
"step": 3690
},
{
"epoch": 17.79,
"learning_rate": 4.111298076923077e-05,
"loss": 0.0141,
"step": 3700
},
{
"epoch": 17.84,
"learning_rate": 4.108894230769231e-05,
"loss": 0.0131,
"step": 3710
},
{
"epoch": 17.88,
"learning_rate": 4.1064903846153844e-05,
"loss": 0.011,
"step": 3720
},
{
"epoch": 17.93,
"learning_rate": 4.104086538461539e-05,
"loss": 0.0113,
"step": 3730
},
{
"epoch": 17.98,
"learning_rate": 4.1016826923076924e-05,
"loss": 0.0124,
"step": 3740
},
{
"epoch": 18.0,
"eval_loss": 0.036256324499845505,
"eval_runtime": 37.0725,
"eval_samples_per_second": 17.506,
"eval_steps_per_second": 2.212,
"step": 3744
},
{
"epoch": 18.03,
"learning_rate": 4.099278846153846e-05,
"loss": 0.0102,
"step": 3750
},
{
"epoch": 18.08,
"learning_rate": 4.096875e-05,
"loss": 0.0132,
"step": 3760
},
{
"epoch": 18.12,
"learning_rate": 4.094471153846154e-05,
"loss": 0.0107,
"step": 3770
},
{
"epoch": 18.17,
"learning_rate": 4.0920673076923076e-05,
"loss": 0.01,
"step": 3780
},
{
"epoch": 18.22,
"learning_rate": 4.089663461538462e-05,
"loss": 0.0119,
"step": 3790
},
{
"epoch": 18.27,
"learning_rate": 4.0872596153846155e-05,
"loss": 0.0115,
"step": 3800
},
{
"epoch": 18.32,
"learning_rate": 4.084855769230769e-05,
"loss": 0.0111,
"step": 3810
},
{
"epoch": 18.37,
"learning_rate": 4.0824519230769235e-05,
"loss": 0.013,
"step": 3820
},
{
"epoch": 18.41,
"learning_rate": 4.080048076923077e-05,
"loss": 0.0093,
"step": 3830
},
{
"epoch": 18.46,
"learning_rate": 4.077644230769231e-05,
"loss": 0.0108,
"step": 3840
},
{
"epoch": 18.51,
"learning_rate": 4.075240384615385e-05,
"loss": 0.0111,
"step": 3850
},
{
"epoch": 18.56,
"learning_rate": 4.0728365384615387e-05,
"loss": 0.0093,
"step": 3860
},
{
"epoch": 18.61,
"learning_rate": 4.070432692307692e-05,
"loss": 0.0111,
"step": 3870
},
{
"epoch": 18.65,
"learning_rate": 4.0680288461538466e-05,
"loss": 0.0124,
"step": 3880
},
{
"epoch": 18.7,
"learning_rate": 4.065625e-05,
"loss": 0.0126,
"step": 3890
},
{
"epoch": 18.75,
"learning_rate": 4.063221153846154e-05,
"loss": 0.0137,
"step": 3900
},
{
"epoch": 18.8,
"learning_rate": 4.060817307692308e-05,
"loss": 0.013,
"step": 3910
},
{
"epoch": 18.85,
"learning_rate": 4.058413461538462e-05,
"loss": 0.0106,
"step": 3920
},
{
"epoch": 18.89,
"learning_rate": 4.0562500000000003e-05,
"loss": 0.0139,
"step": 3930
},
{
"epoch": 18.94,
"learning_rate": 4.053846153846154e-05,
"loss": 0.0144,
"step": 3940
},
{
"epoch": 18.99,
"learning_rate": 4.051442307692308e-05,
"loss": 0.0124,
"step": 3950
},
{
"epoch": 19.0,
"eval_loss": 0.03690133988857269,
"eval_runtime": 36.3716,
"eval_samples_per_second": 17.844,
"eval_steps_per_second": 2.255,
"step": 3952
},
{
"epoch": 19.04,
"learning_rate": 4.049038461538462e-05,
"loss": 0.0086,
"step": 3960
},
{
"epoch": 19.09,
"learning_rate": 4.0466346153846155e-05,
"loss": 0.0082,
"step": 3970
},
{
"epoch": 19.13,
"learning_rate": 4.04423076923077e-05,
"loss": 0.013,
"step": 3980
},
{
"epoch": 19.18,
"learning_rate": 4.0418269230769235e-05,
"loss": 0.0104,
"step": 3990
},
{
"epoch": 19.23,
"learning_rate": 4.039423076923077e-05,
"loss": 0.0083,
"step": 4000
},
{
"epoch": 19.28,
"learning_rate": 4.0370192307692314e-05,
"loss": 0.0122,
"step": 4010
},
{
"epoch": 19.33,
"learning_rate": 4.034615384615385e-05,
"loss": 0.0082,
"step": 4020
},
{
"epoch": 19.38,
"learning_rate": 4.032211538461539e-05,
"loss": 0.0107,
"step": 4030
},
{
"epoch": 19.42,
"learning_rate": 4.029807692307693e-05,
"loss": 0.0063,
"step": 4040
},
{
"epoch": 19.47,
"learning_rate": 4.0274038461538466e-05,
"loss": 0.0081,
"step": 4050
},
{
"epoch": 19.52,
"learning_rate": 4.025e-05,
"loss": 0.0078,
"step": 4060
},
{
"epoch": 19.57,
"learning_rate": 4.0225961538461546e-05,
"loss": 0.0159,
"step": 4070
},
{
"epoch": 19.62,
"learning_rate": 4.020192307692308e-05,
"loss": 0.0099,
"step": 4080
},
{
"epoch": 19.66,
"learning_rate": 4.017788461538462e-05,
"loss": 0.0128,
"step": 4090
},
{
"epoch": 19.71,
"learning_rate": 4.0153846153846155e-05,
"loss": 0.0109,
"step": 4100
},
{
"epoch": 19.76,
"learning_rate": 4.012980769230769e-05,
"loss": 0.0122,
"step": 4110
},
{
"epoch": 19.81,
"learning_rate": 4.010576923076923e-05,
"loss": 0.01,
"step": 4120
},
{
"epoch": 19.86,
"learning_rate": 4.008173076923077e-05,
"loss": 0.0146,
"step": 4130
},
{
"epoch": 19.9,
"learning_rate": 4.005769230769231e-05,
"loss": 0.0099,
"step": 4140
},
{
"epoch": 19.95,
"learning_rate": 4.003365384615384e-05,
"loss": 0.0096,
"step": 4150
},
{
"epoch": 20.0,
"learning_rate": 4.0009615384615386e-05,
"loss": 0.0151,
"step": 4160
},
{
"epoch": 20.0,
"eval_loss": 0.03828124701976776,
"eval_runtime": 36.5852,
"eval_samples_per_second": 17.739,
"eval_steps_per_second": 2.241,
"step": 4160
},
{
"epoch": 20.05,
"learning_rate": 3.998557692307692e-05,
"loss": 0.0074,
"step": 4170
},
{
"epoch": 20.1,
"learning_rate": 3.996153846153846e-05,
"loss": 0.0118,
"step": 4180
},
{
"epoch": 20.14,
"learning_rate": 3.99375e-05,
"loss": 0.0101,
"step": 4190
},
{
"epoch": 20.19,
"learning_rate": 3.991346153846154e-05,
"loss": 0.0089,
"step": 4200
},
{
"epoch": 20.24,
"learning_rate": 3.9889423076923074e-05,
"loss": 0.0102,
"step": 4210
},
{
"epoch": 20.29,
"learning_rate": 3.986538461538462e-05,
"loss": 0.0116,
"step": 4220
},
{
"epoch": 20.34,
"learning_rate": 3.9841346153846154e-05,
"loss": 0.0092,
"step": 4230
},
{
"epoch": 20.38,
"learning_rate": 3.981730769230769e-05,
"loss": 0.0083,
"step": 4240
},
{
"epoch": 20.43,
"learning_rate": 3.979326923076923e-05,
"loss": 0.0107,
"step": 4250
},
{
"epoch": 20.48,
"learning_rate": 3.976923076923077e-05,
"loss": 0.0078,
"step": 4260
},
{
"epoch": 20.53,
"learning_rate": 3.9745192307692306e-05,
"loss": 0.0101,
"step": 4270
},
{
"epoch": 20.58,
"learning_rate": 3.972115384615385e-05,
"loss": 0.0075,
"step": 4280
},
{
"epoch": 20.62,
"learning_rate": 3.9697115384615385e-05,
"loss": 0.0103,
"step": 4290
},
{
"epoch": 20.67,
"learning_rate": 3.967307692307692e-05,
"loss": 0.012,
"step": 4300
},
{
"epoch": 20.72,
"learning_rate": 3.9649038461538465e-05,
"loss": 0.0123,
"step": 4310
},
{
"epoch": 20.77,
"learning_rate": 3.9625e-05,
"loss": 0.007,
"step": 4320
},
{
"epoch": 20.82,
"learning_rate": 3.960096153846154e-05,
"loss": 0.009,
"step": 4330
},
{
"epoch": 20.87,
"learning_rate": 3.957692307692308e-05,
"loss": 0.0071,
"step": 4340
},
{
"epoch": 20.91,
"learning_rate": 3.955288461538462e-05,
"loss": 0.0079,
"step": 4350
},
{
"epoch": 20.96,
"learning_rate": 3.952884615384615e-05,
"loss": 0.0097,
"step": 4360
},
{
"epoch": 21.0,
"eval_loss": 0.04113055393099785,
"eval_runtime": 36.7735,
"eval_samples_per_second": 17.649,
"eval_steps_per_second": 2.23,
"step": 4368
},
{
"epoch": 21.01,
"learning_rate": 3.9504807692307696e-05,
"loss": 0.0104,
"step": 4370
},
{
"epoch": 21.06,
"learning_rate": 3.948076923076923e-05,
"loss": 0.0086,
"step": 4380
},
{
"epoch": 21.11,
"learning_rate": 3.945673076923077e-05,
"loss": 0.0127,
"step": 4390
},
{
"epoch": 21.15,
"learning_rate": 3.943269230769231e-05,
"loss": 0.008,
"step": 4400
},
{
"epoch": 21.2,
"learning_rate": 3.940865384615385e-05,
"loss": 0.0073,
"step": 4410
},
{
"epoch": 21.25,
"learning_rate": 3.9384615384615384e-05,
"loss": 0.0087,
"step": 4420
},
{
"epoch": 21.3,
"learning_rate": 3.936057692307693e-05,
"loss": 0.0089,
"step": 4430
},
{
"epoch": 21.35,
"learning_rate": 3.9336538461538464e-05,
"loss": 0.0097,
"step": 4440
},
{
"epoch": 21.39,
"learning_rate": 3.93125e-05,
"loss": 0.0054,
"step": 4450
},
{
"epoch": 21.44,
"learning_rate": 3.928846153846154e-05,
"loss": 0.0079,
"step": 4460
},
{
"epoch": 21.49,
"learning_rate": 3.926442307692308e-05,
"loss": 0.009,
"step": 4470
},
{
"epoch": 21.54,
"learning_rate": 3.9240384615384616e-05,
"loss": 0.01,
"step": 4480
},
{
"epoch": 21.59,
"learning_rate": 3.921634615384616e-05,
"loss": 0.0085,
"step": 4490
},
{
"epoch": 21.63,
"learning_rate": 3.9192307692307695e-05,
"loss": 0.0094,
"step": 4500
},
{
"epoch": 21.68,
"learning_rate": 3.916826923076923e-05,
"loss": 0.0128,
"step": 4510
},
{
"epoch": 21.73,
"learning_rate": 3.9144230769230775e-05,
"loss": 0.0077,
"step": 4520
},
{
"epoch": 21.78,
"learning_rate": 3.912019230769231e-05,
"loss": 0.0073,
"step": 4530
},
{
"epoch": 21.83,
"learning_rate": 3.909615384615385e-05,
"loss": 0.0083,
"step": 4540
},
{
"epoch": 21.88,
"learning_rate": 3.907211538461539e-05,
"loss": 0.0112,
"step": 4550
},
{
"epoch": 21.92,
"learning_rate": 3.904807692307693e-05,
"loss": 0.0076,
"step": 4560
},
{
"epoch": 21.97,
"learning_rate": 3.902403846153846e-05,
"loss": 0.0084,
"step": 4570
},
{
"epoch": 22.0,
"eval_loss": 0.03989782929420471,
"eval_runtime": 36.902,
"eval_samples_per_second": 17.587,
"eval_steps_per_second": 2.222,
"step": 4576
},
{
"epoch": 22.0,
"step": 4576,
"total_flos": 5.233965363865186e+18,
"train_loss": 0.02212856354686868,
"train_runtime": 7388.5591,
"train_samples_per_second": 22.521,
"train_steps_per_second": 2.815
}
],
"max_steps": 20800,
"num_train_epochs": 100,
"total_flos": 5.233965363865186e+18,
"trial_name": null,
"trial_params": null
}