visual-llm / trainer_state.json
Franklin0314
initial commit
c64c305
raw
history blame
48.2 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.3016596160104132,
"eval_steps": 500,
"global_step": 4000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.9875000000000006e-05,
"loss": 4.8021,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 4.975e-05,
"loss": 3.4946,
"step": 20
},
{
"epoch": 0.01,
"learning_rate": 4.962500000000001e-05,
"loss": 4.2365,
"step": 30
},
{
"epoch": 0.01,
"learning_rate": 4.9500000000000004e-05,
"loss": 3.11,
"step": 40
},
{
"epoch": 0.02,
"learning_rate": 4.937500000000001e-05,
"loss": 2.2186,
"step": 50
},
{
"epoch": 0.02,
"learning_rate": 4.9250000000000004e-05,
"loss": 1.5203,
"step": 60
},
{
"epoch": 0.02,
"learning_rate": 4.9125e-05,
"loss": 1.203,
"step": 70
},
{
"epoch": 0.03,
"learning_rate": 4.9e-05,
"loss": 0.9516,
"step": 80
},
{
"epoch": 0.03,
"learning_rate": 4.8875e-05,
"loss": 0.6212,
"step": 90
},
{
"epoch": 0.03,
"learning_rate": 4.875e-05,
"loss": 0.7066,
"step": 100
},
{
"epoch": 0.04,
"learning_rate": 4.8625e-05,
"loss": 0.2459,
"step": 110
},
{
"epoch": 0.04,
"learning_rate": 4.85e-05,
"loss": 0.4349,
"step": 120
},
{
"epoch": 0.04,
"learning_rate": 4.8375000000000004e-05,
"loss": 0.2229,
"step": 130
},
{
"epoch": 0.05,
"learning_rate": 4.825e-05,
"loss": 0.6896,
"step": 140
},
{
"epoch": 0.05,
"learning_rate": 4.8125000000000004e-05,
"loss": 0.3415,
"step": 150
},
{
"epoch": 0.05,
"learning_rate": 4.8e-05,
"loss": 0.2647,
"step": 160
},
{
"epoch": 0.06,
"learning_rate": 4.7875000000000005e-05,
"loss": 0.3146,
"step": 170
},
{
"epoch": 0.06,
"learning_rate": 4.775e-05,
"loss": 0.259,
"step": 180
},
{
"epoch": 0.06,
"learning_rate": 4.7625000000000006e-05,
"loss": 0.3262,
"step": 190
},
{
"epoch": 0.07,
"learning_rate": 4.75e-05,
"loss": 0.2983,
"step": 200
},
{
"epoch": 0.07,
"learning_rate": 4.7375e-05,
"loss": 0.3234,
"step": 210
},
{
"epoch": 0.07,
"learning_rate": 4.7249999999999997e-05,
"loss": 0.3344,
"step": 220
},
{
"epoch": 0.07,
"learning_rate": 4.7125e-05,
"loss": 0.3257,
"step": 230
},
{
"epoch": 0.08,
"learning_rate": 4.7e-05,
"loss": 0.3293,
"step": 240
},
{
"epoch": 0.08,
"learning_rate": 4.6875e-05,
"loss": 0.1742,
"step": 250
},
{
"epoch": 0.08,
"learning_rate": 4.6750000000000005e-05,
"loss": 0.4563,
"step": 260
},
{
"epoch": 0.09,
"learning_rate": 4.6625e-05,
"loss": 0.2687,
"step": 270
},
{
"epoch": 0.09,
"learning_rate": 4.6500000000000005e-05,
"loss": 0.267,
"step": 280
},
{
"epoch": 0.09,
"learning_rate": 4.6375e-05,
"loss": 0.4485,
"step": 290
},
{
"epoch": 0.1,
"learning_rate": 4.6250000000000006e-05,
"loss": 0.2024,
"step": 300
},
{
"epoch": 0.1,
"learning_rate": 4.6125e-05,
"loss": 0.2136,
"step": 310
},
{
"epoch": 0.1,
"learning_rate": 4.600000000000001e-05,
"loss": 0.2776,
"step": 320
},
{
"epoch": 0.11,
"learning_rate": 4.5875000000000004e-05,
"loss": 0.3112,
"step": 330
},
{
"epoch": 0.11,
"learning_rate": 4.575e-05,
"loss": 0.1997,
"step": 340
},
{
"epoch": 0.11,
"learning_rate": 4.5625e-05,
"loss": 0.2204,
"step": 350
},
{
"epoch": 0.12,
"learning_rate": 4.55e-05,
"loss": 0.2537,
"step": 360
},
{
"epoch": 0.12,
"learning_rate": 4.5375e-05,
"loss": 0.2956,
"step": 370
},
{
"epoch": 0.12,
"learning_rate": 4.525e-05,
"loss": 0.3163,
"step": 380
},
{
"epoch": 0.13,
"learning_rate": 4.5125e-05,
"loss": 0.3704,
"step": 390
},
{
"epoch": 0.13,
"learning_rate": 4.5e-05,
"loss": 0.308,
"step": 400
},
{
"epoch": 0.13,
"learning_rate": 4.4875e-05,
"loss": 0.2247,
"step": 410
},
{
"epoch": 0.14,
"learning_rate": 4.4750000000000004e-05,
"loss": 0.2089,
"step": 420
},
{
"epoch": 0.14,
"learning_rate": 4.4625e-05,
"loss": 0.2159,
"step": 430
},
{
"epoch": 0.14,
"learning_rate": 4.4500000000000004e-05,
"loss": 0.253,
"step": 440
},
{
"epoch": 0.15,
"learning_rate": 4.4375e-05,
"loss": 0.0905,
"step": 450
},
{
"epoch": 0.15,
"learning_rate": 4.4250000000000005e-05,
"loss": 0.3106,
"step": 460
},
{
"epoch": 0.15,
"learning_rate": 4.4125e-05,
"loss": 0.1635,
"step": 470
},
{
"epoch": 0.16,
"learning_rate": 4.4000000000000006e-05,
"loss": 0.2771,
"step": 480
},
{
"epoch": 0.16,
"learning_rate": 4.3875e-05,
"loss": 0.193,
"step": 490
},
{
"epoch": 0.16,
"learning_rate": 4.375e-05,
"loss": 0.3506,
"step": 500
},
{
"epoch": 0.16,
"eval_bleu-4": 0.5123334214062311,
"eval_rouge-1": 74.69764,
"eval_rouge-2": 56.37476000000001,
"eval_rouge-l": 71.65215,
"eval_runtime": 4.9156,
"eval_samples_per_second": 10.172,
"eval_steps_per_second": 0.814,
"step": 500
},
{
"epoch": 0.17,
"learning_rate": 4.3625e-05,
"loss": 0.2192,
"step": 510
},
{
"epoch": 0.17,
"learning_rate": 4.35e-05,
"loss": 0.3669,
"step": 520
},
{
"epoch": 0.17,
"learning_rate": 4.3375000000000004e-05,
"loss": 0.1304,
"step": 530
},
{
"epoch": 0.18,
"learning_rate": 4.325e-05,
"loss": 0.3289,
"step": 540
},
{
"epoch": 0.18,
"learning_rate": 4.3125000000000005e-05,
"loss": 0.3846,
"step": 550
},
{
"epoch": 0.18,
"learning_rate": 4.3e-05,
"loss": 0.2141,
"step": 560
},
{
"epoch": 0.19,
"learning_rate": 4.2875000000000005e-05,
"loss": 0.1857,
"step": 570
},
{
"epoch": 0.19,
"learning_rate": 4.275e-05,
"loss": 0.2101,
"step": 580
},
{
"epoch": 0.19,
"learning_rate": 4.2625000000000006e-05,
"loss": 0.2363,
"step": 590
},
{
"epoch": 0.2,
"learning_rate": 4.25e-05,
"loss": 0.2324,
"step": 600
},
{
"epoch": 0.2,
"learning_rate": 4.237500000000001e-05,
"loss": 0.1082,
"step": 610
},
{
"epoch": 0.2,
"learning_rate": 4.2250000000000004e-05,
"loss": 0.2095,
"step": 620
},
{
"epoch": 0.21,
"learning_rate": 4.2125e-05,
"loss": 0.3072,
"step": 630
},
{
"epoch": 0.21,
"learning_rate": 4.2e-05,
"loss": 0.2708,
"step": 640
},
{
"epoch": 0.21,
"learning_rate": 4.1875e-05,
"loss": 0.1592,
"step": 650
},
{
"epoch": 0.21,
"learning_rate": 4.175e-05,
"loss": 0.3118,
"step": 660
},
{
"epoch": 0.22,
"learning_rate": 4.1625e-05,
"loss": 0.5071,
"step": 670
},
{
"epoch": 0.22,
"learning_rate": 4.15e-05,
"loss": 0.3046,
"step": 680
},
{
"epoch": 0.22,
"learning_rate": 4.1375e-05,
"loss": 0.0869,
"step": 690
},
{
"epoch": 0.23,
"learning_rate": 4.125e-05,
"loss": 0.1031,
"step": 700
},
{
"epoch": 0.23,
"learning_rate": 4.1125000000000004e-05,
"loss": 0.4183,
"step": 710
},
{
"epoch": 0.23,
"learning_rate": 4.1e-05,
"loss": 0.1242,
"step": 720
},
{
"epoch": 0.24,
"learning_rate": 4.0875000000000004e-05,
"loss": 0.2465,
"step": 730
},
{
"epoch": 0.24,
"learning_rate": 4.075e-05,
"loss": 0.1623,
"step": 740
},
{
"epoch": 0.24,
"learning_rate": 4.0625000000000005e-05,
"loss": 0.2403,
"step": 750
},
{
"epoch": 0.25,
"learning_rate": 4.05e-05,
"loss": 0.3134,
"step": 760
},
{
"epoch": 0.25,
"learning_rate": 4.0375e-05,
"loss": 0.2647,
"step": 770
},
{
"epoch": 0.25,
"learning_rate": 4.025e-05,
"loss": 0.2368,
"step": 780
},
{
"epoch": 0.26,
"learning_rate": 4.0125e-05,
"loss": 0.2985,
"step": 790
},
{
"epoch": 0.26,
"learning_rate": 4e-05,
"loss": 0.2929,
"step": 800
},
{
"epoch": 0.26,
"learning_rate": 3.9875e-05,
"loss": 0.2982,
"step": 810
},
{
"epoch": 0.27,
"learning_rate": 3.9750000000000004e-05,
"loss": 0.2775,
"step": 820
},
{
"epoch": 0.27,
"learning_rate": 3.9625e-05,
"loss": 0.3314,
"step": 830
},
{
"epoch": 0.27,
"learning_rate": 3.9500000000000005e-05,
"loss": 0.1401,
"step": 840
},
{
"epoch": 0.28,
"learning_rate": 3.9375e-05,
"loss": 0.2998,
"step": 850
},
{
"epoch": 0.28,
"learning_rate": 3.9250000000000005e-05,
"loss": 0.2041,
"step": 860
},
{
"epoch": 0.28,
"learning_rate": 3.9125e-05,
"loss": 0.3031,
"step": 870
},
{
"epoch": 0.29,
"learning_rate": 3.9000000000000006e-05,
"loss": 0.3725,
"step": 880
},
{
"epoch": 0.29,
"learning_rate": 3.8875e-05,
"loss": 0.1381,
"step": 890
},
{
"epoch": 0.29,
"learning_rate": 3.875e-05,
"loss": 0.166,
"step": 900
},
{
"epoch": 0.3,
"learning_rate": 3.8625e-05,
"loss": 0.242,
"step": 910
},
{
"epoch": 0.3,
"learning_rate": 3.85e-05,
"loss": 0.2182,
"step": 920
},
{
"epoch": 0.3,
"learning_rate": 3.8375e-05,
"loss": 0.2386,
"step": 930
},
{
"epoch": 0.31,
"learning_rate": 3.825e-05,
"loss": 0.0829,
"step": 940
},
{
"epoch": 0.31,
"learning_rate": 3.8125e-05,
"loss": 0.1273,
"step": 950
},
{
"epoch": 0.31,
"learning_rate": 3.8e-05,
"loss": 0.2222,
"step": 960
},
{
"epoch": 0.32,
"learning_rate": 3.7875e-05,
"loss": 0.2664,
"step": 970
},
{
"epoch": 0.32,
"learning_rate": 3.775e-05,
"loss": 0.1874,
"step": 980
},
{
"epoch": 0.32,
"learning_rate": 3.7625e-05,
"loss": 0.1742,
"step": 990
},
{
"epoch": 0.33,
"learning_rate": 3.7500000000000003e-05,
"loss": 0.222,
"step": 1000
},
{
"epoch": 0.33,
"eval_bleu-4": 0.5308189442308736,
"eval_rouge-1": 73.44375,
"eval_rouge-2": 54.7447,
"eval_rouge-l": 71.83562,
"eval_runtime": 4.1772,
"eval_samples_per_second": 11.97,
"eval_steps_per_second": 0.958,
"step": 1000
},
{
"epoch": 0.33,
"learning_rate": 3.737500000000001e-05,
"loss": 0.2489,
"step": 1010
},
{
"epoch": 0.33,
"learning_rate": 3.7250000000000004e-05,
"loss": 0.2511,
"step": 1020
},
{
"epoch": 0.34,
"learning_rate": 3.7125e-05,
"loss": 0.2092,
"step": 1030
},
{
"epoch": 0.34,
"learning_rate": 3.7e-05,
"loss": 0.1308,
"step": 1040
},
{
"epoch": 0.34,
"learning_rate": 3.6875e-05,
"loss": 0.2085,
"step": 1050
},
{
"epoch": 0.34,
"learning_rate": 3.675e-05,
"loss": 0.226,
"step": 1060
},
{
"epoch": 0.35,
"learning_rate": 3.6625e-05,
"loss": 0.252,
"step": 1070
},
{
"epoch": 0.35,
"learning_rate": 3.65e-05,
"loss": 0.122,
"step": 1080
},
{
"epoch": 0.35,
"learning_rate": 3.6375e-05,
"loss": 0.1906,
"step": 1090
},
{
"epoch": 0.36,
"learning_rate": 3.625e-05,
"loss": 0.1968,
"step": 1100
},
{
"epoch": 0.36,
"learning_rate": 3.6125000000000004e-05,
"loss": 0.1916,
"step": 1110
},
{
"epoch": 0.36,
"learning_rate": 3.6e-05,
"loss": 0.4047,
"step": 1120
},
{
"epoch": 0.37,
"learning_rate": 3.5875000000000005e-05,
"loss": 0.1004,
"step": 1130
},
{
"epoch": 0.37,
"learning_rate": 3.575e-05,
"loss": 0.1582,
"step": 1140
},
{
"epoch": 0.37,
"learning_rate": 3.5625000000000005e-05,
"loss": 0.1458,
"step": 1150
},
{
"epoch": 0.38,
"learning_rate": 3.55e-05,
"loss": 0.2366,
"step": 1160
},
{
"epoch": 0.38,
"learning_rate": 3.5375e-05,
"loss": 0.3445,
"step": 1170
},
{
"epoch": 0.38,
"learning_rate": 3.525e-05,
"loss": 0.1609,
"step": 1180
},
{
"epoch": 0.39,
"learning_rate": 3.5125e-05,
"loss": 0.2032,
"step": 1190
},
{
"epoch": 0.39,
"learning_rate": 3.5e-05,
"loss": 0.2324,
"step": 1200
},
{
"epoch": 0.39,
"learning_rate": 3.4875e-05,
"loss": 0.2468,
"step": 1210
},
{
"epoch": 0.4,
"learning_rate": 3.475e-05,
"loss": 0.304,
"step": 1220
},
{
"epoch": 0.4,
"learning_rate": 3.4625e-05,
"loss": 0.1962,
"step": 1230
},
{
"epoch": 0.4,
"learning_rate": 3.45e-05,
"loss": 0.2164,
"step": 1240
},
{
"epoch": 0.41,
"learning_rate": 3.4375e-05,
"loss": 0.3032,
"step": 1250
},
{
"epoch": 0.41,
"learning_rate": 3.4250000000000006e-05,
"loss": 0.1629,
"step": 1260
},
{
"epoch": 0.41,
"learning_rate": 3.4125e-05,
"loss": 0.1357,
"step": 1270
},
{
"epoch": 0.42,
"learning_rate": 3.4000000000000007e-05,
"loss": 0.1634,
"step": 1280
},
{
"epoch": 0.42,
"learning_rate": 3.3875000000000003e-05,
"loss": 0.1235,
"step": 1290
},
{
"epoch": 0.42,
"learning_rate": 3.375000000000001e-05,
"loss": 0.2326,
"step": 1300
},
{
"epoch": 0.43,
"learning_rate": 3.3625000000000004e-05,
"loss": 0.1841,
"step": 1310
},
{
"epoch": 0.43,
"learning_rate": 3.35e-05,
"loss": 0.1866,
"step": 1320
},
{
"epoch": 0.43,
"learning_rate": 3.3375e-05,
"loss": 0.3667,
"step": 1330
},
{
"epoch": 0.44,
"learning_rate": 3.325e-05,
"loss": 0.2589,
"step": 1340
},
{
"epoch": 0.44,
"learning_rate": 3.3125e-05,
"loss": 0.0811,
"step": 1350
},
{
"epoch": 0.44,
"learning_rate": 3.3e-05,
"loss": 0.1739,
"step": 1360
},
{
"epoch": 0.45,
"learning_rate": 3.2875e-05,
"loss": 0.2833,
"step": 1370
},
{
"epoch": 0.45,
"learning_rate": 3.275e-05,
"loss": 0.1666,
"step": 1380
},
{
"epoch": 0.45,
"learning_rate": 3.2625e-05,
"loss": 0.272,
"step": 1390
},
{
"epoch": 0.46,
"learning_rate": 3.2500000000000004e-05,
"loss": 0.2286,
"step": 1400
},
{
"epoch": 0.46,
"learning_rate": 3.2375e-05,
"loss": 0.1464,
"step": 1410
},
{
"epoch": 0.46,
"learning_rate": 3.2250000000000005e-05,
"loss": 0.2486,
"step": 1420
},
{
"epoch": 0.47,
"learning_rate": 3.2125e-05,
"loss": 0.0797,
"step": 1430
},
{
"epoch": 0.47,
"learning_rate": 3.2000000000000005e-05,
"loss": 0.2126,
"step": 1440
},
{
"epoch": 0.47,
"learning_rate": 3.1875e-05,
"loss": 0.2081,
"step": 1450
},
{
"epoch": 0.48,
"learning_rate": 3.175e-05,
"loss": 0.2769,
"step": 1460
},
{
"epoch": 0.48,
"learning_rate": 3.1624999999999996e-05,
"loss": 0.1976,
"step": 1470
},
{
"epoch": 0.48,
"learning_rate": 3.15e-05,
"loss": 0.1582,
"step": 1480
},
{
"epoch": 0.48,
"learning_rate": 3.1375e-05,
"loss": 0.1035,
"step": 1490
},
{
"epoch": 0.49,
"learning_rate": 3.125e-05,
"loss": 0.1899,
"step": 1500
},
{
"epoch": 0.49,
"eval_bleu-4": 0.5282664478029883,
"eval_rouge-1": 73.29392000000001,
"eval_rouge-2": 55.69114,
"eval_rouge-l": 71.72303,
"eval_runtime": 4.6282,
"eval_samples_per_second": 10.803,
"eval_steps_per_second": 0.864,
"step": 1500
},
{
"epoch": 0.49,
"learning_rate": 3.1125000000000004e-05,
"loss": 0.167,
"step": 1510
},
{
"epoch": 0.49,
"learning_rate": 3.1e-05,
"loss": 0.1767,
"step": 1520
},
{
"epoch": 0.5,
"learning_rate": 3.0875000000000005e-05,
"loss": 0.2311,
"step": 1530
},
{
"epoch": 0.5,
"learning_rate": 3.075e-05,
"loss": 0.2156,
"step": 1540
},
{
"epoch": 0.5,
"learning_rate": 3.0625000000000006e-05,
"loss": 0.1911,
"step": 1550
},
{
"epoch": 0.51,
"learning_rate": 3.05e-05,
"loss": 0.1312,
"step": 1560
},
{
"epoch": 0.51,
"learning_rate": 3.0375000000000003e-05,
"loss": 0.1461,
"step": 1570
},
{
"epoch": 0.51,
"learning_rate": 3.025e-05,
"loss": 0.2821,
"step": 1580
},
{
"epoch": 0.52,
"learning_rate": 3.0125000000000004e-05,
"loss": 0.1689,
"step": 1590
},
{
"epoch": 0.52,
"learning_rate": 3e-05,
"loss": 0.1499,
"step": 1600
},
{
"epoch": 0.52,
"learning_rate": 2.9875000000000004e-05,
"loss": 0.2317,
"step": 1610
},
{
"epoch": 0.53,
"learning_rate": 2.975e-05,
"loss": 0.1639,
"step": 1620
},
{
"epoch": 0.53,
"learning_rate": 2.9625000000000002e-05,
"loss": 0.275,
"step": 1630
},
{
"epoch": 0.53,
"learning_rate": 2.95e-05,
"loss": 0.2628,
"step": 1640
},
{
"epoch": 0.54,
"learning_rate": 2.9375000000000003e-05,
"loss": 0.1617,
"step": 1650
},
{
"epoch": 0.54,
"learning_rate": 2.925e-05,
"loss": 0.303,
"step": 1660
},
{
"epoch": 0.54,
"learning_rate": 2.9125000000000003e-05,
"loss": 0.0989,
"step": 1670
},
{
"epoch": 0.55,
"learning_rate": 2.9e-05,
"loss": 0.0732,
"step": 1680
},
{
"epoch": 0.55,
"learning_rate": 2.8875e-05,
"loss": 0.2459,
"step": 1690
},
{
"epoch": 0.55,
"learning_rate": 2.8749999999999997e-05,
"loss": 0.2137,
"step": 1700
},
{
"epoch": 0.56,
"learning_rate": 2.8625e-05,
"loss": 0.2293,
"step": 1710
},
{
"epoch": 0.56,
"learning_rate": 2.8499999999999998e-05,
"loss": 0.1635,
"step": 1720
},
{
"epoch": 0.56,
"learning_rate": 2.8375000000000002e-05,
"loss": 0.1743,
"step": 1730
},
{
"epoch": 0.57,
"learning_rate": 2.825e-05,
"loss": 0.1878,
"step": 1740
},
{
"epoch": 0.57,
"learning_rate": 2.8125000000000003e-05,
"loss": 0.2032,
"step": 1750
},
{
"epoch": 0.57,
"learning_rate": 2.8000000000000003e-05,
"loss": 0.2278,
"step": 1760
},
{
"epoch": 0.58,
"learning_rate": 2.7875e-05,
"loss": 0.2181,
"step": 1770
},
{
"epoch": 0.58,
"learning_rate": 2.7750000000000004e-05,
"loss": 0.17,
"step": 1780
},
{
"epoch": 0.58,
"learning_rate": 2.7625e-05,
"loss": 0.2624,
"step": 1790
},
{
"epoch": 0.59,
"learning_rate": 2.7500000000000004e-05,
"loss": 0.2703,
"step": 1800
},
{
"epoch": 0.59,
"learning_rate": 2.7375e-05,
"loss": 0.27,
"step": 1810
},
{
"epoch": 0.59,
"learning_rate": 2.725e-05,
"loss": 0.1659,
"step": 1820
},
{
"epoch": 0.6,
"learning_rate": 2.7125000000000002e-05,
"loss": 0.1295,
"step": 1830
},
{
"epoch": 0.6,
"learning_rate": 2.7000000000000002e-05,
"loss": 0.3632,
"step": 1840
},
{
"epoch": 0.6,
"learning_rate": 2.6875e-05,
"loss": 0.1305,
"step": 1850
},
{
"epoch": 0.61,
"learning_rate": 2.6750000000000003e-05,
"loss": 0.287,
"step": 1860
},
{
"epoch": 0.61,
"learning_rate": 2.6625e-05,
"loss": 0.1507,
"step": 1870
},
{
"epoch": 0.61,
"learning_rate": 2.6500000000000004e-05,
"loss": 0.0894,
"step": 1880
},
{
"epoch": 0.62,
"learning_rate": 2.6375e-05,
"loss": 0.1896,
"step": 1890
},
{
"epoch": 0.62,
"learning_rate": 2.625e-05,
"loss": 0.1462,
"step": 1900
},
{
"epoch": 0.62,
"learning_rate": 2.6124999999999998e-05,
"loss": 0.1062,
"step": 1910
},
{
"epoch": 0.62,
"learning_rate": 2.6000000000000002e-05,
"loss": 0.1156,
"step": 1920
},
{
"epoch": 0.63,
"learning_rate": 2.5875e-05,
"loss": 0.1539,
"step": 1930
},
{
"epoch": 0.63,
"learning_rate": 2.5750000000000002e-05,
"loss": 0.0697,
"step": 1940
},
{
"epoch": 0.63,
"learning_rate": 2.5625e-05,
"loss": 0.1109,
"step": 1950
},
{
"epoch": 0.64,
"learning_rate": 2.5500000000000003e-05,
"loss": 0.1212,
"step": 1960
},
{
"epoch": 0.64,
"learning_rate": 2.5375e-05,
"loss": 0.1903,
"step": 1970
},
{
"epoch": 0.64,
"learning_rate": 2.525e-05,
"loss": 0.1561,
"step": 1980
},
{
"epoch": 0.65,
"learning_rate": 2.5124999999999997e-05,
"loss": 0.3047,
"step": 1990
},
{
"epoch": 0.65,
"learning_rate": 2.5e-05,
"loss": 0.3121,
"step": 2000
},
{
"epoch": 0.65,
"eval_bleu-4": 0.5122339544955379,
"eval_rouge-1": 72.57773,
"eval_rouge-2": 52.838330000000006,
"eval_rouge-l": 70.44752,
"eval_runtime": 4.1957,
"eval_samples_per_second": 11.917,
"eval_steps_per_second": 0.953,
"step": 2000
},
{
"epoch": 0.65,
"learning_rate": 2.4875e-05,
"loss": 0.1308,
"step": 2010
},
{
"epoch": 0.66,
"learning_rate": 2.4750000000000002e-05,
"loss": 0.241,
"step": 2020
},
{
"epoch": 0.66,
"learning_rate": 2.4625000000000002e-05,
"loss": 0.2604,
"step": 2030
},
{
"epoch": 0.66,
"learning_rate": 2.45e-05,
"loss": 0.178,
"step": 2040
},
{
"epoch": 0.67,
"learning_rate": 2.4375e-05,
"loss": 0.2828,
"step": 2050
},
{
"epoch": 0.67,
"learning_rate": 2.425e-05,
"loss": 0.3039,
"step": 2060
},
{
"epoch": 0.67,
"learning_rate": 2.4125e-05,
"loss": 0.1629,
"step": 2070
},
{
"epoch": 0.68,
"learning_rate": 2.4e-05,
"loss": 0.2903,
"step": 2080
},
{
"epoch": 0.68,
"learning_rate": 2.3875e-05,
"loss": 0.2469,
"step": 2090
},
{
"epoch": 0.68,
"learning_rate": 2.375e-05,
"loss": 0.1666,
"step": 2100
},
{
"epoch": 0.69,
"learning_rate": 2.3624999999999998e-05,
"loss": 0.1642,
"step": 2110
},
{
"epoch": 0.69,
"learning_rate": 2.35e-05,
"loss": 0.1691,
"step": 2120
},
{
"epoch": 0.69,
"learning_rate": 2.3375000000000002e-05,
"loss": 0.1123,
"step": 2130
},
{
"epoch": 0.7,
"learning_rate": 2.3250000000000003e-05,
"loss": 0.2387,
"step": 2140
},
{
"epoch": 0.7,
"learning_rate": 2.3125000000000003e-05,
"loss": 0.1778,
"step": 2150
},
{
"epoch": 0.7,
"learning_rate": 2.3000000000000003e-05,
"loss": 0.1903,
"step": 2160
},
{
"epoch": 0.71,
"learning_rate": 2.2875e-05,
"loss": 0.161,
"step": 2170
},
{
"epoch": 0.71,
"learning_rate": 2.275e-05,
"loss": 0.2142,
"step": 2180
},
{
"epoch": 0.71,
"learning_rate": 2.2625e-05,
"loss": 0.1828,
"step": 2190
},
{
"epoch": 0.72,
"learning_rate": 2.25e-05,
"loss": 0.1311,
"step": 2200
},
{
"epoch": 0.72,
"learning_rate": 2.2375000000000002e-05,
"loss": 0.361,
"step": 2210
},
{
"epoch": 0.72,
"learning_rate": 2.2250000000000002e-05,
"loss": 0.1018,
"step": 2220
},
{
"epoch": 0.73,
"learning_rate": 2.2125000000000002e-05,
"loss": 0.2022,
"step": 2230
},
{
"epoch": 0.73,
"learning_rate": 2.2000000000000003e-05,
"loss": 0.2202,
"step": 2240
},
{
"epoch": 0.73,
"learning_rate": 2.1875e-05,
"loss": 0.1076,
"step": 2250
},
{
"epoch": 0.74,
"learning_rate": 2.175e-05,
"loss": 0.1287,
"step": 2260
},
{
"epoch": 0.74,
"learning_rate": 2.1625e-05,
"loss": 0.126,
"step": 2270
},
{
"epoch": 0.74,
"learning_rate": 2.15e-05,
"loss": 0.2802,
"step": 2280
},
{
"epoch": 0.75,
"learning_rate": 2.1375e-05,
"loss": 0.2538,
"step": 2290
},
{
"epoch": 0.75,
"learning_rate": 2.125e-05,
"loss": 0.1371,
"step": 2300
},
{
"epoch": 0.75,
"learning_rate": 2.1125000000000002e-05,
"loss": 0.1423,
"step": 2310
},
{
"epoch": 0.75,
"learning_rate": 2.1e-05,
"loss": 0.1213,
"step": 2320
},
{
"epoch": 0.76,
"learning_rate": 2.0875e-05,
"loss": 0.0962,
"step": 2330
},
{
"epoch": 0.76,
"learning_rate": 2.075e-05,
"loss": 0.1903,
"step": 2340
},
{
"epoch": 0.76,
"learning_rate": 2.0625e-05,
"loss": 0.25,
"step": 2350
},
{
"epoch": 0.77,
"learning_rate": 2.05e-05,
"loss": 0.1043,
"step": 2360
},
{
"epoch": 0.77,
"learning_rate": 2.0375e-05,
"loss": 0.2189,
"step": 2370
},
{
"epoch": 0.77,
"learning_rate": 2.025e-05,
"loss": 0.2064,
"step": 2380
},
{
"epoch": 0.78,
"learning_rate": 2.0125e-05,
"loss": 0.2896,
"step": 2390
},
{
"epoch": 0.78,
"learning_rate": 2e-05,
"loss": 0.1534,
"step": 2400
},
{
"epoch": 0.78,
"learning_rate": 1.9875000000000002e-05,
"loss": 0.152,
"step": 2410
},
{
"epoch": 0.79,
"learning_rate": 1.9750000000000002e-05,
"loss": 0.1111,
"step": 2420
},
{
"epoch": 0.79,
"learning_rate": 1.9625000000000003e-05,
"loss": 0.2323,
"step": 2430
},
{
"epoch": 0.79,
"learning_rate": 1.9500000000000003e-05,
"loss": 0.2582,
"step": 2440
},
{
"epoch": 0.8,
"learning_rate": 1.9375e-05,
"loss": 0.2148,
"step": 2450
},
{
"epoch": 0.8,
"learning_rate": 1.925e-05,
"loss": 0.2708,
"step": 2460
},
{
"epoch": 0.8,
"learning_rate": 1.9125e-05,
"loss": 0.1505,
"step": 2470
},
{
"epoch": 0.81,
"learning_rate": 1.9e-05,
"loss": 0.1196,
"step": 2480
},
{
"epoch": 0.81,
"learning_rate": 1.8875e-05,
"loss": 0.2677,
"step": 2490
},
{
"epoch": 0.81,
"learning_rate": 1.8750000000000002e-05,
"loss": 0.1215,
"step": 2500
},
{
"epoch": 0.81,
"eval_bleu-4": 0.5122156520784964,
"eval_rouge-1": 74.73468000000001,
"eval_rouge-2": 53.852579999999996,
"eval_rouge-l": 72.49319,
"eval_runtime": 4.2013,
"eval_samples_per_second": 11.901,
"eval_steps_per_second": 0.952,
"step": 2500
},
{
"epoch": 0.82,
"learning_rate": 1.8625000000000002e-05,
"loss": 0.1761,
"step": 2510
},
{
"epoch": 0.82,
"learning_rate": 1.85e-05,
"loss": 0.1645,
"step": 2520
},
{
"epoch": 0.82,
"learning_rate": 1.8375e-05,
"loss": 0.1746,
"step": 2530
},
{
"epoch": 0.83,
"learning_rate": 1.825e-05,
"loss": 0.2182,
"step": 2540
},
{
"epoch": 0.83,
"learning_rate": 1.8125e-05,
"loss": 0.2265,
"step": 2550
},
{
"epoch": 0.83,
"learning_rate": 1.8e-05,
"loss": 0.2604,
"step": 2560
},
{
"epoch": 0.84,
"learning_rate": 1.7875e-05,
"loss": 0.2326,
"step": 2570
},
{
"epoch": 0.84,
"learning_rate": 1.775e-05,
"loss": 0.1621,
"step": 2580
},
{
"epoch": 0.84,
"learning_rate": 1.7625e-05,
"loss": 0.1838,
"step": 2590
},
{
"epoch": 0.85,
"learning_rate": 1.75e-05,
"loss": 0.1687,
"step": 2600
},
{
"epoch": 0.85,
"learning_rate": 1.7375e-05,
"loss": 0.0636,
"step": 2610
},
{
"epoch": 0.85,
"learning_rate": 1.725e-05,
"loss": 0.0831,
"step": 2620
},
{
"epoch": 0.86,
"learning_rate": 1.7125000000000003e-05,
"loss": 0.2057,
"step": 2630
},
{
"epoch": 0.86,
"learning_rate": 1.7000000000000003e-05,
"loss": 0.1846,
"step": 2640
},
{
"epoch": 0.86,
"learning_rate": 1.6875000000000004e-05,
"loss": 0.1197,
"step": 2650
},
{
"epoch": 0.87,
"learning_rate": 1.675e-05,
"loss": 0.245,
"step": 2660
},
{
"epoch": 0.87,
"learning_rate": 1.6625e-05,
"loss": 0.1317,
"step": 2670
},
{
"epoch": 0.87,
"learning_rate": 1.65e-05,
"loss": 0.2095,
"step": 2680
},
{
"epoch": 0.88,
"learning_rate": 1.6375e-05,
"loss": 0.11,
"step": 2690
},
{
"epoch": 0.88,
"learning_rate": 1.6250000000000002e-05,
"loss": 0.1467,
"step": 2700
},
{
"epoch": 0.88,
"learning_rate": 1.6125000000000002e-05,
"loss": 0.3093,
"step": 2710
},
{
"epoch": 0.89,
"learning_rate": 1.6000000000000003e-05,
"loss": 0.1817,
"step": 2720
},
{
"epoch": 0.89,
"learning_rate": 1.5875e-05,
"loss": 0.2466,
"step": 2730
},
{
"epoch": 0.89,
"learning_rate": 1.575e-05,
"loss": 0.1661,
"step": 2740
},
{
"epoch": 0.89,
"learning_rate": 1.5625e-05,
"loss": 0.2242,
"step": 2750
},
{
"epoch": 0.9,
"learning_rate": 1.55e-05,
"loss": 0.2207,
"step": 2760
},
{
"epoch": 0.9,
"learning_rate": 1.5375e-05,
"loss": 0.1961,
"step": 2770
},
{
"epoch": 0.9,
"learning_rate": 1.525e-05,
"loss": 0.2673,
"step": 2780
},
{
"epoch": 0.91,
"learning_rate": 1.5125e-05,
"loss": 0.1608,
"step": 2790
},
{
"epoch": 0.91,
"learning_rate": 1.5e-05,
"loss": 0.0949,
"step": 2800
},
{
"epoch": 0.91,
"learning_rate": 1.4875e-05,
"loss": 0.1448,
"step": 2810
},
{
"epoch": 0.92,
"learning_rate": 1.475e-05,
"loss": 0.2233,
"step": 2820
},
{
"epoch": 0.92,
"learning_rate": 1.4625e-05,
"loss": 0.2646,
"step": 2830
},
{
"epoch": 0.92,
"learning_rate": 1.45e-05,
"loss": 0.1473,
"step": 2840
},
{
"epoch": 0.93,
"learning_rate": 1.4374999999999999e-05,
"loss": 0.1528,
"step": 2850
},
{
"epoch": 0.93,
"learning_rate": 1.4249999999999999e-05,
"loss": 0.3152,
"step": 2860
},
{
"epoch": 0.93,
"learning_rate": 1.4125e-05,
"loss": 0.138,
"step": 2870
},
{
"epoch": 0.94,
"learning_rate": 1.4000000000000001e-05,
"loss": 0.0929,
"step": 2880
},
{
"epoch": 0.94,
"learning_rate": 1.3875000000000002e-05,
"loss": 0.1869,
"step": 2890
},
{
"epoch": 0.94,
"learning_rate": 1.3750000000000002e-05,
"loss": 0.2122,
"step": 2900
},
{
"epoch": 0.95,
"learning_rate": 1.3625e-05,
"loss": 0.2408,
"step": 2910
},
{
"epoch": 0.95,
"learning_rate": 1.3500000000000001e-05,
"loss": 0.3734,
"step": 2920
},
{
"epoch": 0.95,
"learning_rate": 1.3375000000000002e-05,
"loss": 0.3088,
"step": 2930
},
{
"epoch": 0.96,
"learning_rate": 1.3250000000000002e-05,
"loss": 0.1843,
"step": 2940
},
{
"epoch": 0.96,
"learning_rate": 1.3125e-05,
"loss": 0.2538,
"step": 2950
},
{
"epoch": 0.96,
"learning_rate": 1.3000000000000001e-05,
"loss": 0.1594,
"step": 2960
},
{
"epoch": 0.97,
"learning_rate": 1.2875000000000001e-05,
"loss": 0.1556,
"step": 2970
},
{
"epoch": 0.97,
"learning_rate": 1.2750000000000002e-05,
"loss": 0.1398,
"step": 2980
},
{
"epoch": 0.97,
"learning_rate": 1.2625e-05,
"loss": 0.2144,
"step": 2990
},
{
"epoch": 0.98,
"learning_rate": 1.25e-05,
"loss": 0.1546,
"step": 3000
},
{
"epoch": 0.98,
"eval_bleu-4": 0.5104840369798801,
"eval_rouge-1": 73.24027000000001,
"eval_rouge-2": 54.12907,
"eval_rouge-l": 71.3942,
"eval_runtime": 4.0442,
"eval_samples_per_second": 12.363,
"eval_steps_per_second": 0.989,
"step": 3000
},
{
"epoch": 0.98,
"learning_rate": 1.2375000000000001e-05,
"loss": 0.2081,
"step": 3010
},
{
"epoch": 0.98,
"learning_rate": 1.225e-05,
"loss": 0.2799,
"step": 3020
},
{
"epoch": 0.99,
"learning_rate": 1.2125e-05,
"loss": 0.1146,
"step": 3030
},
{
"epoch": 0.99,
"learning_rate": 1.2e-05,
"loss": 0.1702,
"step": 3040
},
{
"epoch": 0.99,
"learning_rate": 1.1875e-05,
"loss": 0.1742,
"step": 3050
},
{
"epoch": 1.0,
"learning_rate": 1.175e-05,
"loss": 0.1785,
"step": 3060
},
{
"epoch": 1.0,
"learning_rate": 1.1625000000000001e-05,
"loss": 0.1969,
"step": 3070
},
{
"epoch": 1.0,
"learning_rate": 1.1500000000000002e-05,
"loss": 0.1249,
"step": 3080
},
{
"epoch": 1.01,
"learning_rate": 1.1375e-05,
"loss": 0.0861,
"step": 3090
},
{
"epoch": 1.01,
"learning_rate": 1.125e-05,
"loss": 0.2002,
"step": 3100
},
{
"epoch": 1.01,
"learning_rate": 1.1125000000000001e-05,
"loss": 0.1078,
"step": 3110
},
{
"epoch": 1.02,
"learning_rate": 1.1000000000000001e-05,
"loss": 0.1899,
"step": 3120
},
{
"epoch": 1.02,
"learning_rate": 1.0875e-05,
"loss": 0.1088,
"step": 3130
},
{
"epoch": 1.02,
"learning_rate": 1.075e-05,
"loss": 0.1926,
"step": 3140
},
{
"epoch": 1.03,
"learning_rate": 1.0625e-05,
"loss": 0.1961,
"step": 3150
},
{
"epoch": 1.03,
"learning_rate": 1.05e-05,
"loss": 0.2562,
"step": 3160
},
{
"epoch": 1.03,
"learning_rate": 1.0375e-05,
"loss": 0.0961,
"step": 3170
},
{
"epoch": 1.03,
"learning_rate": 1.025e-05,
"loss": 0.1666,
"step": 3180
},
{
"epoch": 1.04,
"learning_rate": 1.0125e-05,
"loss": 0.2314,
"step": 3190
},
{
"epoch": 1.04,
"learning_rate": 1e-05,
"loss": 0.2981,
"step": 3200
},
{
"epoch": 1.04,
"learning_rate": 9.875000000000001e-06,
"loss": 0.1123,
"step": 3210
},
{
"epoch": 1.05,
"learning_rate": 9.750000000000002e-06,
"loss": 0.1554,
"step": 3220
},
{
"epoch": 1.05,
"learning_rate": 9.625e-06,
"loss": 0.2253,
"step": 3230
},
{
"epoch": 1.05,
"learning_rate": 9.5e-06,
"loss": 0.1191,
"step": 3240
},
{
"epoch": 1.06,
"learning_rate": 9.375000000000001e-06,
"loss": 0.1305,
"step": 3250
},
{
"epoch": 1.06,
"learning_rate": 9.25e-06,
"loss": 0.1272,
"step": 3260
},
{
"epoch": 1.06,
"learning_rate": 9.125e-06,
"loss": 0.1163,
"step": 3270
},
{
"epoch": 1.07,
"learning_rate": 9e-06,
"loss": 0.1584,
"step": 3280
},
{
"epoch": 1.07,
"learning_rate": 8.875e-06,
"loss": 0.0484,
"step": 3290
},
{
"epoch": 1.07,
"learning_rate": 8.75e-06,
"loss": 0.192,
"step": 3300
},
{
"epoch": 1.08,
"learning_rate": 8.625e-06,
"loss": 0.1396,
"step": 3310
},
{
"epoch": 1.08,
"learning_rate": 8.500000000000002e-06,
"loss": 0.2213,
"step": 3320
},
{
"epoch": 1.08,
"learning_rate": 8.375e-06,
"loss": 0.1782,
"step": 3330
},
{
"epoch": 1.09,
"learning_rate": 8.25e-06,
"loss": 0.2035,
"step": 3340
},
{
"epoch": 1.09,
"learning_rate": 8.125000000000001e-06,
"loss": 0.281,
"step": 3350
},
{
"epoch": 1.09,
"learning_rate": 8.000000000000001e-06,
"loss": 0.0522,
"step": 3360
},
{
"epoch": 1.1,
"learning_rate": 7.875e-06,
"loss": 0.1353,
"step": 3370
},
{
"epoch": 1.1,
"learning_rate": 7.75e-06,
"loss": 0.081,
"step": 3380
},
{
"epoch": 1.1,
"learning_rate": 7.625e-06,
"loss": 0.1356,
"step": 3390
},
{
"epoch": 1.11,
"learning_rate": 7.5e-06,
"loss": 0.1515,
"step": 3400
},
{
"epoch": 1.11,
"learning_rate": 7.375e-06,
"loss": 0.1786,
"step": 3410
},
{
"epoch": 1.11,
"learning_rate": 7.25e-06,
"loss": 0.0977,
"step": 3420
},
{
"epoch": 1.12,
"learning_rate": 7.1249999999999995e-06,
"loss": 0.3118,
"step": 3430
},
{
"epoch": 1.12,
"learning_rate": 7.000000000000001e-06,
"loss": 0.2391,
"step": 3440
},
{
"epoch": 1.12,
"learning_rate": 6.875000000000001e-06,
"loss": 0.2432,
"step": 3450
},
{
"epoch": 1.13,
"learning_rate": 6.750000000000001e-06,
"loss": 0.2216,
"step": 3460
},
{
"epoch": 1.13,
"learning_rate": 6.625000000000001e-06,
"loss": 0.1875,
"step": 3470
},
{
"epoch": 1.13,
"learning_rate": 6.5000000000000004e-06,
"loss": 0.1712,
"step": 3480
},
{
"epoch": 1.14,
"learning_rate": 6.375000000000001e-06,
"loss": 0.1565,
"step": 3490
},
{
"epoch": 1.14,
"learning_rate": 6.25e-06,
"loss": 0.2192,
"step": 3500
},
{
"epoch": 1.14,
"eval_bleu-4": 0.4973012355868603,
"eval_rouge-1": 73.60392,
"eval_rouge-2": 51.70972,
"eval_rouge-l": 70.33331,
"eval_runtime": 4.1995,
"eval_samples_per_second": 11.906,
"eval_steps_per_second": 0.952,
"step": 3500
},
{
"epoch": 1.14,
"learning_rate": 6.125e-06,
"loss": 0.0656,
"step": 3510
},
{
"epoch": 1.15,
"learning_rate": 6e-06,
"loss": 0.1279,
"step": 3520
},
{
"epoch": 1.15,
"learning_rate": 5.875e-06,
"loss": 0.2456,
"step": 3530
},
{
"epoch": 1.15,
"learning_rate": 5.750000000000001e-06,
"loss": 0.1141,
"step": 3540
},
{
"epoch": 1.16,
"learning_rate": 5.625e-06,
"loss": 0.1019,
"step": 3550
},
{
"epoch": 1.16,
"learning_rate": 5.500000000000001e-06,
"loss": 0.1476,
"step": 3560
},
{
"epoch": 1.16,
"learning_rate": 5.375e-06,
"loss": 0.1546,
"step": 3570
},
{
"epoch": 1.16,
"learning_rate": 5.25e-06,
"loss": 0.0519,
"step": 3580
},
{
"epoch": 1.17,
"learning_rate": 5.125e-06,
"loss": 0.0889,
"step": 3590
},
{
"epoch": 1.17,
"learning_rate": 5e-06,
"loss": 0.1611,
"step": 3600
},
{
"epoch": 1.17,
"learning_rate": 4.875000000000001e-06,
"loss": 0.1526,
"step": 3610
},
{
"epoch": 1.18,
"learning_rate": 4.75e-06,
"loss": 0.1407,
"step": 3620
},
{
"epoch": 1.18,
"learning_rate": 4.625e-06,
"loss": 0.1389,
"step": 3630
},
{
"epoch": 1.18,
"learning_rate": 4.5e-06,
"loss": 0.1123,
"step": 3640
},
{
"epoch": 1.19,
"learning_rate": 4.375e-06,
"loss": 0.1315,
"step": 3650
},
{
"epoch": 1.19,
"learning_rate": 4.250000000000001e-06,
"loss": 0.1828,
"step": 3660
},
{
"epoch": 1.19,
"learning_rate": 4.125e-06,
"loss": 0.0759,
"step": 3670
},
{
"epoch": 1.2,
"learning_rate": 4.000000000000001e-06,
"loss": 0.1373,
"step": 3680
},
{
"epoch": 1.2,
"learning_rate": 3.875e-06,
"loss": 0.0871,
"step": 3690
},
{
"epoch": 1.2,
"learning_rate": 3.75e-06,
"loss": 0.1816,
"step": 3700
},
{
"epoch": 1.21,
"learning_rate": 3.625e-06,
"loss": 0.1257,
"step": 3710
},
{
"epoch": 1.21,
"learning_rate": 3.5000000000000004e-06,
"loss": 0.152,
"step": 3720
},
{
"epoch": 1.21,
"learning_rate": 3.3750000000000003e-06,
"loss": 0.2231,
"step": 3730
},
{
"epoch": 1.22,
"learning_rate": 3.2500000000000002e-06,
"loss": 0.4279,
"step": 3740
},
{
"epoch": 1.22,
"learning_rate": 3.125e-06,
"loss": 0.1672,
"step": 3750
},
{
"epoch": 1.22,
"learning_rate": 3e-06,
"loss": 0.1343,
"step": 3760
},
{
"epoch": 1.23,
"learning_rate": 2.8750000000000004e-06,
"loss": 0.1901,
"step": 3770
},
{
"epoch": 1.23,
"learning_rate": 2.7500000000000004e-06,
"loss": 0.2439,
"step": 3780
},
{
"epoch": 1.23,
"learning_rate": 2.625e-06,
"loss": 0.196,
"step": 3790
},
{
"epoch": 1.24,
"learning_rate": 2.5e-06,
"loss": 0.2084,
"step": 3800
},
{
"epoch": 1.24,
"learning_rate": 2.375e-06,
"loss": 0.3074,
"step": 3810
},
{
"epoch": 1.24,
"learning_rate": 2.25e-06,
"loss": 0.117,
"step": 3820
},
{
"epoch": 1.25,
"learning_rate": 2.1250000000000004e-06,
"loss": 0.1888,
"step": 3830
},
{
"epoch": 1.25,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.1905,
"step": 3840
},
{
"epoch": 1.25,
"learning_rate": 1.875e-06,
"loss": 0.1035,
"step": 3850
},
{
"epoch": 1.26,
"learning_rate": 1.7500000000000002e-06,
"loss": 0.1589,
"step": 3860
},
{
"epoch": 1.26,
"learning_rate": 1.6250000000000001e-06,
"loss": 0.1249,
"step": 3870
},
{
"epoch": 1.26,
"learning_rate": 1.5e-06,
"loss": 0.1453,
"step": 3880
},
{
"epoch": 1.27,
"learning_rate": 1.3750000000000002e-06,
"loss": 0.1559,
"step": 3890
},
{
"epoch": 1.27,
"learning_rate": 1.25e-06,
"loss": 0.0999,
"step": 3900
},
{
"epoch": 1.27,
"learning_rate": 1.125e-06,
"loss": 0.1109,
"step": 3910
},
{
"epoch": 1.28,
"learning_rate": 1.0000000000000002e-06,
"loss": 0.1272,
"step": 3920
},
{
"epoch": 1.28,
"learning_rate": 8.750000000000001e-07,
"loss": 0.1684,
"step": 3930
},
{
"epoch": 1.28,
"learning_rate": 7.5e-07,
"loss": 0.0721,
"step": 3940
},
{
"epoch": 1.29,
"learning_rate": 6.25e-07,
"loss": 0.0791,
"step": 3950
},
{
"epoch": 1.29,
"learning_rate": 5.000000000000001e-07,
"loss": 0.169,
"step": 3960
},
{
"epoch": 1.29,
"learning_rate": 3.75e-07,
"loss": 0.1407,
"step": 3970
},
{
"epoch": 1.3,
"learning_rate": 2.5000000000000004e-07,
"loss": 0.1715,
"step": 3980
},
{
"epoch": 1.3,
"learning_rate": 1.2500000000000002e-07,
"loss": 0.1107,
"step": 3990
},
{
"epoch": 1.3,
"learning_rate": 0.0,
"loss": 0.1829,
"step": 4000
},
{
"epoch": 1.3,
"eval_bleu-4": 0.5180444909128761,
"eval_rouge-1": 74.50813,
"eval_rouge-2": 52.175689999999996,
"eval_rouge-l": 71.20649999999999,
"eval_runtime": 4.1977,
"eval_samples_per_second": 11.911,
"eval_steps_per_second": 0.953,
"step": 4000
}
],
"logging_steps": 10,
"max_steps": 4000,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"total_flos": 1.0020007388258304e+16,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}