mT5-small-wmt2014-calm-ft / trainer_state.json
Sehoon's picture
store
417d881
raw
history blame
122 kB
{
"best_metric": 14.437,
"best_model_checkpoint": "ckpt_mt5_calm/google/mt5-small/wmt14_de_en/lr0.0005_e5/checkpoint-500000",
"epoch": 1.8137594282459693,
"global_step": 510000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 0.0005,
"loss": 5.1585,
"step": 500
},
{
"epoch": 0.0,
"learning_rate": 0.0005,
"loss": 3.9783,
"step": 1000
},
{
"epoch": 0.01,
"learning_rate": 0.0005,
"loss": 3.7381,
"step": 1500
},
{
"epoch": 0.01,
"learning_rate": 0.0005,
"loss": 3.605,
"step": 2000
},
{
"epoch": 0.01,
"learning_rate": 0.0005,
"loss": 3.4827,
"step": 2500
},
{
"epoch": 0.01,
"learning_rate": 0.0005,
"loss": 3.4053,
"step": 3000
},
{
"epoch": 0.01,
"learning_rate": 0.0005,
"loss": 3.363,
"step": 3500
},
{
"epoch": 0.01,
"learning_rate": 0.0005,
"loss": 3.2694,
"step": 4000
},
{
"epoch": 0.02,
"learning_rate": 0.0005,
"loss": 3.1942,
"step": 4500
},
{
"epoch": 0.02,
"learning_rate": 0.0005,
"loss": 3.1715,
"step": 5000
},
{
"epoch": 0.02,
"learning_rate": 0.0005,
"loss": 3.1428,
"step": 5500
},
{
"epoch": 0.02,
"learning_rate": 0.0005,
"loss": 3.1076,
"step": 6000
},
{
"epoch": 0.02,
"learning_rate": 0.0005,
"loss": 3.0584,
"step": 6500
},
{
"epoch": 0.02,
"learning_rate": 0.0005,
"loss": 3.0355,
"step": 7000
},
{
"epoch": 0.03,
"learning_rate": 0.0005,
"loss": 3.0121,
"step": 7500
},
{
"epoch": 0.03,
"learning_rate": 0.0005,
"loss": 2.9444,
"step": 8000
},
{
"epoch": 0.03,
"learning_rate": 0.0005,
"loss": 2.9554,
"step": 8500
},
{
"epoch": 0.03,
"learning_rate": 0.0005,
"loss": 2.9541,
"step": 9000
},
{
"epoch": 0.03,
"learning_rate": 0.0005,
"loss": 2.9076,
"step": 9500
},
{
"epoch": 0.04,
"learning_rate": 0.0005,
"loss": 2.9165,
"step": 10000
},
{
"epoch": 0.04,
"eval_bleu": 9.5968,
"eval_gen_len": 17.2947,
"eval_runtime": 210.2888,
"eval_samples_per_second": 14.266,
"eval_steps_per_second": 1.783,
"step": 10000
},
{
"epoch": 0.04,
"learning_rate": 0.0005,
"loss": 2.9083,
"step": 10500
},
{
"epoch": 0.04,
"learning_rate": 0.0005,
"loss": 2.8379,
"step": 11000
},
{
"epoch": 0.04,
"learning_rate": 0.0005,
"loss": 2.8713,
"step": 11500
},
{
"epoch": 0.04,
"learning_rate": 0.0005,
"loss": 2.8174,
"step": 12000
},
{
"epoch": 0.04,
"learning_rate": 0.0005,
"loss": 2.8014,
"step": 12500
},
{
"epoch": 0.05,
"learning_rate": 0.0005,
"loss": 2.8065,
"step": 13000
},
{
"epoch": 0.05,
"learning_rate": 0.0005,
"loss": 2.7795,
"step": 13500
},
{
"epoch": 0.05,
"learning_rate": 0.0005,
"loss": 2.765,
"step": 14000
},
{
"epoch": 0.05,
"learning_rate": 0.0005,
"loss": 2.7456,
"step": 14500
},
{
"epoch": 0.05,
"learning_rate": 0.0005,
"loss": 2.7718,
"step": 15000
},
{
"epoch": 0.06,
"learning_rate": 0.0005,
"loss": 2.7426,
"step": 15500
},
{
"epoch": 0.06,
"learning_rate": 0.0005,
"loss": 2.7086,
"step": 16000
},
{
"epoch": 0.06,
"learning_rate": 0.0005,
"loss": 2.72,
"step": 16500
},
{
"epoch": 0.06,
"learning_rate": 0.0005,
"loss": 2.6916,
"step": 17000
},
{
"epoch": 0.06,
"learning_rate": 0.0005,
"loss": 2.7049,
"step": 17500
},
{
"epoch": 0.06,
"learning_rate": 0.0005,
"loss": 2.7094,
"step": 18000
},
{
"epoch": 0.07,
"learning_rate": 0.0005,
"loss": 2.691,
"step": 18500
},
{
"epoch": 0.07,
"learning_rate": 0.0005,
"loss": 2.7057,
"step": 19000
},
{
"epoch": 0.07,
"learning_rate": 0.0005,
"loss": 2.6476,
"step": 19500
},
{
"epoch": 0.07,
"learning_rate": 0.0005,
"loss": 2.6388,
"step": 20000
},
{
"epoch": 0.07,
"eval_bleu": 10.7385,
"eval_gen_len": 17.1993,
"eval_runtime": 210.0921,
"eval_samples_per_second": 14.279,
"eval_steps_per_second": 1.785,
"step": 20000
},
{
"epoch": 0.07,
"learning_rate": 0.0005,
"loss": 2.6002,
"step": 20500
},
{
"epoch": 0.07,
"learning_rate": 0.0005,
"loss": 2.6225,
"step": 21000
},
{
"epoch": 0.08,
"learning_rate": 0.0005,
"loss": 2.6422,
"step": 21500
},
{
"epoch": 0.08,
"learning_rate": 0.0005,
"loss": 2.6291,
"step": 22000
},
{
"epoch": 0.08,
"learning_rate": 0.0005,
"loss": 2.612,
"step": 22500
},
{
"epoch": 0.08,
"learning_rate": 0.0005,
"loss": 2.5761,
"step": 23000
},
{
"epoch": 0.08,
"learning_rate": 0.0005,
"loss": 2.5875,
"step": 23500
},
{
"epoch": 0.09,
"learning_rate": 0.0005,
"loss": 2.5992,
"step": 24000
},
{
"epoch": 0.09,
"learning_rate": 0.0005,
"loss": 2.5601,
"step": 24500
},
{
"epoch": 0.09,
"learning_rate": 0.0005,
"loss": 2.5567,
"step": 25000
},
{
"epoch": 0.09,
"learning_rate": 0.0005,
"loss": 2.5571,
"step": 25500
},
{
"epoch": 0.09,
"learning_rate": 0.0005,
"loss": 2.5457,
"step": 26000
},
{
"epoch": 0.09,
"learning_rate": 0.0005,
"loss": 2.524,
"step": 26500
},
{
"epoch": 0.1,
"learning_rate": 0.0005,
"loss": 2.545,
"step": 27000
},
{
"epoch": 0.1,
"learning_rate": 0.0005,
"loss": 2.5502,
"step": 27500
},
{
"epoch": 0.1,
"learning_rate": 0.0005,
"loss": 2.5193,
"step": 28000
},
{
"epoch": 0.1,
"learning_rate": 0.0005,
"loss": 2.518,
"step": 28500
},
{
"epoch": 0.1,
"learning_rate": 0.0005,
"loss": 2.5476,
"step": 29000
},
{
"epoch": 0.1,
"learning_rate": 0.0005,
"loss": 2.5378,
"step": 29500
},
{
"epoch": 0.11,
"learning_rate": 0.0005,
"loss": 2.5397,
"step": 30000
},
{
"epoch": 0.11,
"eval_bleu": 11.2222,
"eval_gen_len": 17.2613,
"eval_runtime": 209.8126,
"eval_samples_per_second": 14.298,
"eval_steps_per_second": 1.787,
"step": 30000
},
{
"epoch": 0.11,
"learning_rate": 0.0005,
"loss": 2.4887,
"step": 30500
},
{
"epoch": 0.11,
"learning_rate": 0.0005,
"loss": 2.4931,
"step": 31000
},
{
"epoch": 0.11,
"learning_rate": 0.0005,
"loss": 2.514,
"step": 31500
},
{
"epoch": 0.11,
"learning_rate": 0.0005,
"loss": 2.4727,
"step": 32000
},
{
"epoch": 0.12,
"learning_rate": 0.0005,
"loss": 2.4937,
"step": 32500
},
{
"epoch": 0.12,
"learning_rate": 0.0005,
"loss": 2.4849,
"step": 33000
},
{
"epoch": 0.12,
"learning_rate": 0.0005,
"loss": 2.4951,
"step": 33500
},
{
"epoch": 0.12,
"learning_rate": 0.0005,
"loss": 2.4652,
"step": 34000
},
{
"epoch": 0.12,
"learning_rate": 0.0005,
"loss": 2.4523,
"step": 34500
},
{
"epoch": 0.12,
"learning_rate": 0.0005,
"loss": 2.4862,
"step": 35000
},
{
"epoch": 0.13,
"learning_rate": 0.0005,
"loss": 2.4779,
"step": 35500
},
{
"epoch": 0.13,
"learning_rate": 0.0005,
"loss": 2.4616,
"step": 36000
},
{
"epoch": 0.13,
"learning_rate": 0.0005,
"loss": 2.4679,
"step": 36500
},
{
"epoch": 0.13,
"learning_rate": 0.0005,
"loss": 2.4274,
"step": 37000
},
{
"epoch": 0.13,
"learning_rate": 0.0005,
"loss": 2.4542,
"step": 37500
},
{
"epoch": 0.14,
"learning_rate": 0.0005,
"loss": 2.4711,
"step": 38000
},
{
"epoch": 0.14,
"learning_rate": 0.0005,
"loss": 2.4417,
"step": 38500
},
{
"epoch": 0.14,
"learning_rate": 0.0005,
"loss": 2.444,
"step": 39000
},
{
"epoch": 0.14,
"learning_rate": 0.0005,
"loss": 2.4355,
"step": 39500
},
{
"epoch": 0.14,
"learning_rate": 0.0005,
"loss": 2.4444,
"step": 40000
},
{
"epoch": 0.14,
"eval_bleu": 11.6796,
"eval_gen_len": 17.225,
"eval_runtime": 208.6699,
"eval_samples_per_second": 14.377,
"eval_steps_per_second": 1.797,
"step": 40000
},
{
"epoch": 0.14,
"learning_rate": 0.0005,
"loss": 2.4323,
"step": 40500
},
{
"epoch": 0.15,
"learning_rate": 0.0005,
"loss": 2.4094,
"step": 41000
},
{
"epoch": 0.15,
"learning_rate": 0.0005,
"loss": 2.415,
"step": 41500
},
{
"epoch": 0.15,
"learning_rate": 0.0005,
"loss": 2.3979,
"step": 42000
},
{
"epoch": 0.15,
"learning_rate": 0.0005,
"loss": 2.3854,
"step": 42500
},
{
"epoch": 0.15,
"learning_rate": 0.0005,
"loss": 2.3759,
"step": 43000
},
{
"epoch": 0.15,
"learning_rate": 0.0005,
"loss": 2.3881,
"step": 43500
},
{
"epoch": 0.16,
"learning_rate": 0.0005,
"loss": 2.3981,
"step": 44000
},
{
"epoch": 0.16,
"learning_rate": 0.0005,
"loss": 2.4224,
"step": 44500
},
{
"epoch": 0.16,
"learning_rate": 0.0005,
"loss": 2.3761,
"step": 45000
},
{
"epoch": 0.16,
"learning_rate": 0.0005,
"loss": 2.4399,
"step": 45500
},
{
"epoch": 0.16,
"learning_rate": 0.0005,
"loss": 2.3923,
"step": 46000
},
{
"epoch": 0.17,
"learning_rate": 0.0005,
"loss": 2.4098,
"step": 46500
},
{
"epoch": 0.17,
"learning_rate": 0.0005,
"loss": 2.3789,
"step": 47000
},
{
"epoch": 0.17,
"learning_rate": 0.0005,
"loss": 2.4062,
"step": 47500
},
{
"epoch": 0.17,
"learning_rate": 0.0005,
"loss": 2.3814,
"step": 48000
},
{
"epoch": 0.17,
"learning_rate": 0.0005,
"loss": 2.3801,
"step": 48500
},
{
"epoch": 0.17,
"learning_rate": 0.0005,
"loss": 2.3709,
"step": 49000
},
{
"epoch": 0.18,
"learning_rate": 0.0005,
"loss": 2.3442,
"step": 49500
},
{
"epoch": 0.18,
"learning_rate": 0.0005,
"loss": 2.3634,
"step": 50000
},
{
"epoch": 0.18,
"eval_bleu": 12.1549,
"eval_gen_len": 17.2763,
"eval_runtime": 207.6227,
"eval_samples_per_second": 14.449,
"eval_steps_per_second": 1.806,
"step": 50000
},
{
"epoch": 0.18,
"learning_rate": 0.0005,
"loss": 2.3769,
"step": 50500
},
{
"epoch": 0.18,
"learning_rate": 0.0005,
"loss": 2.3646,
"step": 51000
},
{
"epoch": 0.18,
"learning_rate": 0.0005,
"loss": 2.3628,
"step": 51500
},
{
"epoch": 0.18,
"learning_rate": 0.0005,
"loss": 2.3378,
"step": 52000
},
{
"epoch": 0.19,
"learning_rate": 0.0005,
"loss": 2.3609,
"step": 52500
},
{
"epoch": 0.19,
"learning_rate": 0.0005,
"loss": 2.3754,
"step": 53000
},
{
"epoch": 0.19,
"learning_rate": 0.0005,
"loss": 2.3555,
"step": 53500
},
{
"epoch": 0.19,
"learning_rate": 0.0005,
"loss": 2.3596,
"step": 54000
},
{
"epoch": 0.19,
"learning_rate": 0.0005,
"loss": 2.3222,
"step": 54500
},
{
"epoch": 0.2,
"learning_rate": 0.0005,
"loss": 2.3257,
"step": 55000
},
{
"epoch": 0.2,
"learning_rate": 0.0005,
"loss": 2.3631,
"step": 55500
},
{
"epoch": 0.2,
"learning_rate": 0.0005,
"loss": 2.3317,
"step": 56000
},
{
"epoch": 0.2,
"learning_rate": 0.0005,
"loss": 2.3404,
"step": 56500
},
{
"epoch": 0.2,
"learning_rate": 0.0005,
"loss": 2.3737,
"step": 57000
},
{
"epoch": 0.2,
"learning_rate": 0.0005,
"loss": 2.3384,
"step": 57500
},
{
"epoch": 0.21,
"learning_rate": 0.0005,
"loss": 2.3218,
"step": 58000
},
{
"epoch": 0.21,
"learning_rate": 0.0005,
"loss": 2.3527,
"step": 58500
},
{
"epoch": 0.21,
"learning_rate": 0.0005,
"loss": 2.3268,
"step": 59000
},
{
"epoch": 0.21,
"learning_rate": 0.0005,
"loss": 2.3099,
"step": 59500
},
{
"epoch": 0.21,
"learning_rate": 0.0005,
"loss": 2.3032,
"step": 60000
},
{
"epoch": 0.21,
"eval_bleu": 12.3766,
"eval_gen_len": 17.197,
"eval_runtime": 210.5803,
"eval_samples_per_second": 14.246,
"eval_steps_per_second": 1.781,
"step": 60000
},
{
"epoch": 0.22,
"learning_rate": 0.0005,
"loss": 2.309,
"step": 60500
},
{
"epoch": 0.22,
"learning_rate": 0.0005,
"loss": 2.3257,
"step": 61000
},
{
"epoch": 0.22,
"learning_rate": 0.0005,
"loss": 2.3094,
"step": 61500
},
{
"epoch": 0.22,
"learning_rate": 0.0005,
"loss": 2.2655,
"step": 62000
},
{
"epoch": 0.22,
"learning_rate": 0.0005,
"loss": 2.2845,
"step": 62500
},
{
"epoch": 0.22,
"learning_rate": 0.0005,
"loss": 2.264,
"step": 63000
},
{
"epoch": 0.23,
"learning_rate": 0.0005,
"loss": 2.2825,
"step": 63500
},
{
"epoch": 0.23,
"learning_rate": 0.0005,
"loss": 2.3123,
"step": 64000
},
{
"epoch": 0.23,
"learning_rate": 0.0005,
"loss": 2.3179,
"step": 64500
},
{
"epoch": 0.23,
"learning_rate": 0.0005,
"loss": 2.2915,
"step": 65000
},
{
"epoch": 0.23,
"learning_rate": 0.0005,
"loss": 2.2926,
"step": 65500
},
{
"epoch": 0.23,
"learning_rate": 0.0005,
"loss": 2.2963,
"step": 66000
},
{
"epoch": 0.24,
"learning_rate": 0.0005,
"loss": 2.3041,
"step": 66500
},
{
"epoch": 0.24,
"learning_rate": 0.0005,
"loss": 2.2735,
"step": 67000
},
{
"epoch": 0.24,
"learning_rate": 0.0005,
"loss": 2.2915,
"step": 67500
},
{
"epoch": 0.24,
"learning_rate": 0.0005,
"loss": 2.2724,
"step": 68000
},
{
"epoch": 0.24,
"learning_rate": 0.0005,
"loss": 2.2816,
"step": 68500
},
{
"epoch": 0.25,
"learning_rate": 0.0005,
"loss": 2.2798,
"step": 69000
},
{
"epoch": 0.25,
"learning_rate": 0.0005,
"loss": 2.2817,
"step": 69500
},
{
"epoch": 0.25,
"learning_rate": 0.0005,
"loss": 2.2821,
"step": 70000
},
{
"epoch": 0.25,
"eval_bleu": 0,
"eval_gen_len": 17.236,
"eval_runtime": 205.7096,
"eval_samples_per_second": 14.584,
"eval_steps_per_second": 1.823,
"step": 70000
},
{
"epoch": 0.25,
"learning_rate": 0.0005,
"loss": 2.2764,
"step": 70500
},
{
"epoch": 0.25,
"learning_rate": 0.0005,
"loss": 2.277,
"step": 71000
},
{
"epoch": 0.25,
"learning_rate": 0.0005,
"loss": 2.248,
"step": 71500
},
{
"epoch": 0.26,
"learning_rate": 0.0005,
"loss": 2.2635,
"step": 72000
},
{
"epoch": 0.26,
"learning_rate": 0.0005,
"loss": 2.2778,
"step": 72500
},
{
"epoch": 0.26,
"learning_rate": 0.0005,
"loss": 2.274,
"step": 73000
},
{
"epoch": 0.26,
"learning_rate": 0.0005,
"loss": 2.2447,
"step": 73500
},
{
"epoch": 0.26,
"learning_rate": 0.0005,
"loss": 2.2507,
"step": 74000
},
{
"epoch": 0.26,
"learning_rate": 0.0005,
"loss": 2.2641,
"step": 74500
},
{
"epoch": 0.27,
"learning_rate": 0.0005,
"loss": 2.2654,
"step": 75000
},
{
"epoch": 0.27,
"learning_rate": 0.0005,
"loss": 2.2361,
"step": 75500
},
{
"epoch": 0.27,
"learning_rate": 0.0005,
"loss": 2.2509,
"step": 76000
},
{
"epoch": 0.27,
"learning_rate": 0.0005,
"loss": 2.2538,
"step": 76500
},
{
"epoch": 0.27,
"learning_rate": 0.0005,
"loss": 2.2482,
"step": 77000
},
{
"epoch": 0.28,
"learning_rate": 0.0005,
"loss": 2.2958,
"step": 77500
},
{
"epoch": 0.28,
"learning_rate": 0.0005,
"loss": 2.2403,
"step": 78000
},
{
"epoch": 0.28,
"learning_rate": 0.0005,
"loss": 2.2521,
"step": 78500
},
{
"epoch": 0.28,
"learning_rate": 0.0005,
"loss": 2.2556,
"step": 79000
},
{
"epoch": 0.28,
"learning_rate": 0.0005,
"loss": 2.2235,
"step": 79500
},
{
"epoch": 0.28,
"learning_rate": 0.0005,
"loss": 2.2452,
"step": 80000
},
{
"epoch": 0.28,
"eval_bleu": 12.4778,
"eval_gen_len": 17.1983,
"eval_runtime": 210.1956,
"eval_samples_per_second": 14.272,
"eval_steps_per_second": 1.784,
"step": 80000
},
{
"epoch": 0.29,
"learning_rate": 0.0005,
"loss": 2.2499,
"step": 80500
},
{
"epoch": 0.29,
"learning_rate": 0.0005,
"loss": 2.2461,
"step": 81000
},
{
"epoch": 0.29,
"learning_rate": 0.0005,
"loss": 2.2056,
"step": 81500
},
{
"epoch": 0.29,
"learning_rate": 0.0005,
"loss": 2.2334,
"step": 82000
},
{
"epoch": 0.29,
"learning_rate": 0.0005,
"loss": 2.217,
"step": 82500
},
{
"epoch": 0.3,
"learning_rate": 0.0005,
"loss": 2.2491,
"step": 83000
},
{
"epoch": 0.3,
"learning_rate": 0.0005,
"loss": 2.229,
"step": 83500
},
{
"epoch": 0.3,
"learning_rate": 0.0005,
"loss": 2.2197,
"step": 84000
},
{
"epoch": 0.3,
"learning_rate": 0.0005,
"loss": 2.2652,
"step": 84500
},
{
"epoch": 0.3,
"learning_rate": 0.0005,
"loss": 2.2195,
"step": 85000
},
{
"epoch": 0.3,
"learning_rate": 0.0005,
"loss": 2.2357,
"step": 85500
},
{
"epoch": 0.31,
"learning_rate": 0.0005,
"loss": 2.2303,
"step": 86000
},
{
"epoch": 0.31,
"learning_rate": 0.0005,
"loss": 2.2241,
"step": 86500
},
{
"epoch": 0.31,
"learning_rate": 0.0005,
"loss": 2.2229,
"step": 87000
},
{
"epoch": 0.31,
"learning_rate": 0.0005,
"loss": 2.2029,
"step": 87500
},
{
"epoch": 0.31,
"learning_rate": 0.0005,
"loss": 2.2047,
"step": 88000
},
{
"epoch": 0.31,
"learning_rate": 0.0005,
"loss": 2.2189,
"step": 88500
},
{
"epoch": 0.32,
"learning_rate": 0.0005,
"loss": 2.2186,
"step": 89000
},
{
"epoch": 0.32,
"learning_rate": 0.0005,
"loss": 2.1882,
"step": 89500
},
{
"epoch": 0.32,
"learning_rate": 0.0005,
"loss": 2.2099,
"step": 90000
},
{
"epoch": 0.32,
"eval_bleu": 12.9363,
"eval_gen_len": 17.2027,
"eval_runtime": 211.5928,
"eval_samples_per_second": 14.178,
"eval_steps_per_second": 1.772,
"step": 90000
},
{
"epoch": 0.32,
"learning_rate": 0.0005,
"loss": 2.2175,
"step": 90500
},
{
"epoch": 0.32,
"learning_rate": 0.0005,
"loss": 2.2035,
"step": 91000
},
{
"epoch": 0.33,
"learning_rate": 0.0005,
"loss": 2.2162,
"step": 91500
},
{
"epoch": 0.33,
"learning_rate": 0.0005,
"loss": 2.189,
"step": 92000
},
{
"epoch": 0.33,
"learning_rate": 0.0005,
"loss": 2.2124,
"step": 92500
},
{
"epoch": 0.33,
"learning_rate": 0.0005,
"loss": 2.2191,
"step": 93000
},
{
"epoch": 0.33,
"learning_rate": 0.0005,
"loss": 2.2225,
"step": 93500
},
{
"epoch": 0.33,
"learning_rate": 0.0005,
"loss": 2.1997,
"step": 94000
},
{
"epoch": 0.34,
"learning_rate": 0.0005,
"loss": 2.1941,
"step": 94500
},
{
"epoch": 0.34,
"learning_rate": 0.0005,
"loss": 2.2108,
"step": 95000
},
{
"epoch": 0.34,
"learning_rate": 0.0005,
"loss": 2.1919,
"step": 95500
},
{
"epoch": 0.34,
"learning_rate": 0.0005,
"loss": 2.197,
"step": 96000
},
{
"epoch": 0.34,
"learning_rate": 0.0005,
"loss": 2.2065,
"step": 96500
},
{
"epoch": 0.35,
"learning_rate": 0.0005,
"loss": 2.1878,
"step": 97000
},
{
"epoch": 0.35,
"learning_rate": 0.0005,
"loss": 2.1815,
"step": 97500
},
{
"epoch": 0.35,
"learning_rate": 0.0005,
"loss": 2.2084,
"step": 98000
},
{
"epoch": 0.35,
"learning_rate": 0.0005,
"loss": 2.1968,
"step": 98500
},
{
"epoch": 0.35,
"learning_rate": 0.0005,
"loss": 2.1875,
"step": 99000
},
{
"epoch": 0.35,
"learning_rate": 0.0005,
"loss": 2.1867,
"step": 99500
},
{
"epoch": 0.36,
"learning_rate": 0.0005,
"loss": 2.1648,
"step": 100000
},
{
"epoch": 0.36,
"eval_bleu": 12.6533,
"eval_gen_len": 17.2053,
"eval_runtime": 210.3408,
"eval_samples_per_second": 14.263,
"eval_steps_per_second": 1.783,
"step": 100000
},
{
"epoch": 0.36,
"learning_rate": 0.0005,
"loss": 2.1614,
"step": 100500
},
{
"epoch": 0.36,
"learning_rate": 0.0005,
"loss": 2.1697,
"step": 101000
},
{
"epoch": 0.36,
"learning_rate": 0.0005,
"loss": 2.1723,
"step": 101500
},
{
"epoch": 0.36,
"learning_rate": 0.0005,
"loss": 2.1693,
"step": 102000
},
{
"epoch": 0.36,
"learning_rate": 0.0005,
"loss": 2.1926,
"step": 102500
},
{
"epoch": 0.37,
"learning_rate": 0.0005,
"loss": 2.1633,
"step": 103000
},
{
"epoch": 0.37,
"learning_rate": 0.0005,
"loss": 2.191,
"step": 103500
},
{
"epoch": 0.37,
"learning_rate": 0.0005,
"loss": 2.1545,
"step": 104000
},
{
"epoch": 0.37,
"learning_rate": 0.0005,
"loss": 2.1767,
"step": 104500
},
{
"epoch": 0.37,
"learning_rate": 0.0005,
"loss": 2.1776,
"step": 105000
},
{
"epoch": 0.38,
"learning_rate": 0.0005,
"loss": 2.1842,
"step": 105500
},
{
"epoch": 0.38,
"learning_rate": 0.0005,
"loss": 2.1691,
"step": 106000
},
{
"epoch": 0.38,
"learning_rate": 0.0005,
"loss": 2.1628,
"step": 106500
},
{
"epoch": 0.38,
"learning_rate": 0.0005,
"loss": 2.1709,
"step": 107000
},
{
"epoch": 0.38,
"learning_rate": 0.0005,
"loss": 2.1679,
"step": 107500
},
{
"epoch": 0.38,
"learning_rate": 0.0005,
"loss": 2.1679,
"step": 108000
},
{
"epoch": 0.39,
"learning_rate": 0.0005,
"loss": 2.1696,
"step": 108500
},
{
"epoch": 0.39,
"learning_rate": 0.0005,
"loss": 2.1717,
"step": 109000
},
{
"epoch": 0.39,
"learning_rate": 0.0005,
"loss": 2.1893,
"step": 109500
},
{
"epoch": 0.39,
"learning_rate": 0.0005,
"loss": 2.1695,
"step": 110000
},
{
"epoch": 0.39,
"eval_bleu": 12.8902,
"eval_gen_len": 17.2487,
"eval_runtime": 211.7604,
"eval_samples_per_second": 14.167,
"eval_steps_per_second": 1.771,
"step": 110000
},
{
"epoch": 0.39,
"learning_rate": 0.0005,
"loss": 2.1338,
"step": 110500
},
{
"epoch": 0.39,
"learning_rate": 0.0005,
"loss": 2.2026,
"step": 111000
},
{
"epoch": 0.4,
"learning_rate": 0.0005,
"loss": 2.155,
"step": 111500
},
{
"epoch": 0.4,
"learning_rate": 0.0005,
"loss": 2.1416,
"step": 112000
},
{
"epoch": 0.4,
"learning_rate": 0.0005,
"loss": 2.1372,
"step": 112500
},
{
"epoch": 0.4,
"learning_rate": 0.0005,
"loss": 2.1753,
"step": 113000
},
{
"epoch": 0.4,
"learning_rate": 0.0005,
"loss": 2.1391,
"step": 113500
},
{
"epoch": 0.41,
"learning_rate": 0.0005,
"loss": 2.1305,
"step": 114000
},
{
"epoch": 0.41,
"learning_rate": 0.0005,
"loss": 2.099,
"step": 114500
},
{
"epoch": 0.41,
"learning_rate": 0.0005,
"loss": 2.1598,
"step": 115000
},
{
"epoch": 0.41,
"learning_rate": 0.0005,
"loss": 2.1628,
"step": 115500
},
{
"epoch": 0.41,
"learning_rate": 0.0005,
"loss": 2.1374,
"step": 116000
},
{
"epoch": 0.41,
"learning_rate": 0.0005,
"loss": 2.1699,
"step": 116500
},
{
"epoch": 0.42,
"learning_rate": 0.0005,
"loss": 2.1612,
"step": 117000
},
{
"epoch": 0.42,
"learning_rate": 0.0005,
"loss": 2.1554,
"step": 117500
},
{
"epoch": 0.42,
"learning_rate": 0.0005,
"loss": 2.1399,
"step": 118000
},
{
"epoch": 0.42,
"learning_rate": 0.0005,
"loss": 2.139,
"step": 118500
},
{
"epoch": 0.42,
"learning_rate": 0.0005,
"loss": 2.1397,
"step": 119000
},
{
"epoch": 0.43,
"learning_rate": 0.0005,
"loss": 2.1328,
"step": 119500
},
{
"epoch": 0.43,
"learning_rate": 0.0005,
"loss": 2.1567,
"step": 120000
},
{
"epoch": 0.43,
"eval_bleu": 12.8997,
"eval_gen_len": 17.247,
"eval_runtime": 210.8317,
"eval_samples_per_second": 14.229,
"eval_steps_per_second": 1.779,
"step": 120000
},
{
"epoch": 0.43,
"learning_rate": 0.0005,
"loss": 2.137,
"step": 120500
},
{
"epoch": 0.43,
"learning_rate": 0.0005,
"loss": 2.1359,
"step": 121000
},
{
"epoch": 0.43,
"learning_rate": 0.0005,
"loss": 2.1319,
"step": 121500
},
{
"epoch": 0.43,
"learning_rate": 0.0005,
"loss": 2.1224,
"step": 122000
},
{
"epoch": 0.44,
"learning_rate": 0.0005,
"loss": 2.1555,
"step": 122500
},
{
"epoch": 0.44,
"learning_rate": 0.0005,
"loss": 2.1085,
"step": 123000
},
{
"epoch": 0.44,
"learning_rate": 0.0005,
"loss": 2.1293,
"step": 123500
},
{
"epoch": 0.44,
"learning_rate": 0.0005,
"loss": 2.1237,
"step": 124000
},
{
"epoch": 0.44,
"learning_rate": 0.0005,
"loss": 2.1444,
"step": 124500
},
{
"epoch": 0.44,
"learning_rate": 0.0005,
"loss": 2.1284,
"step": 125000
},
{
"epoch": 0.45,
"learning_rate": 0.0005,
"loss": 2.1433,
"step": 125500
},
{
"epoch": 0.45,
"learning_rate": 0.0005,
"loss": 2.1435,
"step": 126000
},
{
"epoch": 0.45,
"learning_rate": 0.0005,
"loss": 2.1512,
"step": 126500
},
{
"epoch": 0.45,
"learning_rate": 0.0005,
"loss": 2.1306,
"step": 127000
},
{
"epoch": 0.45,
"learning_rate": 0.0005,
"loss": 2.1311,
"step": 127500
},
{
"epoch": 0.46,
"learning_rate": 0.0005,
"loss": 2.1175,
"step": 128000
},
{
"epoch": 0.46,
"learning_rate": 0.0005,
"loss": 2.1288,
"step": 128500
},
{
"epoch": 0.46,
"learning_rate": 0.0005,
"loss": 2.102,
"step": 129000
},
{
"epoch": 0.46,
"learning_rate": 0.0005,
"loss": 2.1171,
"step": 129500
},
{
"epoch": 0.46,
"learning_rate": 0.0005,
"loss": 2.1323,
"step": 130000
},
{
"epoch": 0.46,
"eval_bleu": 13.1654,
"eval_gen_len": 17.275,
"eval_runtime": 209.873,
"eval_samples_per_second": 14.294,
"eval_steps_per_second": 1.787,
"step": 130000
},
{
"epoch": 0.46,
"learning_rate": 0.0005,
"loss": 2.1132,
"step": 130500
},
{
"epoch": 0.47,
"learning_rate": 0.0005,
"loss": 2.1226,
"step": 131000
},
{
"epoch": 0.47,
"learning_rate": 0.0005,
"loss": 2.0993,
"step": 131500
},
{
"epoch": 0.47,
"learning_rate": 0.0005,
"loss": 2.1318,
"step": 132000
},
{
"epoch": 0.47,
"learning_rate": 0.0005,
"loss": 2.1263,
"step": 132500
},
{
"epoch": 0.47,
"learning_rate": 0.0005,
"loss": 2.1316,
"step": 133000
},
{
"epoch": 0.47,
"learning_rate": 0.0005,
"loss": 2.1146,
"step": 133500
},
{
"epoch": 0.48,
"learning_rate": 0.0005,
"loss": 2.1242,
"step": 134000
},
{
"epoch": 0.48,
"learning_rate": 0.0005,
"loss": 2.1184,
"step": 134500
},
{
"epoch": 0.48,
"learning_rate": 0.0005,
"loss": 2.1291,
"step": 135000
},
{
"epoch": 0.48,
"learning_rate": 0.0005,
"loss": 2.0707,
"step": 135500
},
{
"epoch": 0.48,
"learning_rate": 0.0005,
"loss": 2.0974,
"step": 136000
},
{
"epoch": 0.49,
"learning_rate": 0.0005,
"loss": 2.1385,
"step": 136500
},
{
"epoch": 0.49,
"learning_rate": 0.0005,
"loss": 2.1038,
"step": 137000
},
{
"epoch": 0.49,
"learning_rate": 0.0005,
"loss": 2.1058,
"step": 137500
},
{
"epoch": 0.49,
"learning_rate": 0.0005,
"loss": 2.1388,
"step": 138000
},
{
"epoch": 0.49,
"learning_rate": 0.0005,
"loss": 2.1259,
"step": 138500
},
{
"epoch": 0.49,
"learning_rate": 0.0005,
"loss": 2.1081,
"step": 139000
},
{
"epoch": 0.5,
"learning_rate": 0.0005,
"loss": 2.1273,
"step": 139500
},
{
"epoch": 0.5,
"learning_rate": 0.0005,
"loss": 2.0737,
"step": 140000
},
{
"epoch": 0.5,
"eval_bleu": 13.3266,
"eval_gen_len": 17.2593,
"eval_runtime": 210.3807,
"eval_samples_per_second": 14.26,
"eval_steps_per_second": 1.782,
"step": 140000
},
{
"epoch": 0.5,
"learning_rate": 0.0005,
"loss": 2.1144,
"step": 140500
},
{
"epoch": 0.5,
"learning_rate": 0.0005,
"loss": 2.0973,
"step": 141000
},
{
"epoch": 0.5,
"learning_rate": 0.0005,
"loss": 2.1156,
"step": 141500
},
{
"epoch": 0.51,
"learning_rate": 0.0005,
"loss": 2.1344,
"step": 142000
},
{
"epoch": 0.51,
"learning_rate": 0.0005,
"loss": 2.1184,
"step": 142500
},
{
"epoch": 0.51,
"learning_rate": 0.0005,
"loss": 2.1114,
"step": 143000
},
{
"epoch": 0.51,
"learning_rate": 0.0005,
"loss": 2.0971,
"step": 143500
},
{
"epoch": 0.51,
"learning_rate": 0.0005,
"loss": 2.092,
"step": 144000
},
{
"epoch": 0.51,
"learning_rate": 0.0005,
"loss": 2.101,
"step": 144500
},
{
"epoch": 0.52,
"learning_rate": 0.0005,
"loss": 2.0607,
"step": 145000
},
{
"epoch": 0.52,
"learning_rate": 0.0005,
"loss": 2.1177,
"step": 145500
},
{
"epoch": 0.52,
"learning_rate": 0.0005,
"loss": 2.1142,
"step": 146000
},
{
"epoch": 0.52,
"learning_rate": 0.0005,
"loss": 2.1028,
"step": 146500
},
{
"epoch": 0.52,
"learning_rate": 0.0005,
"loss": 2.1253,
"step": 147000
},
{
"epoch": 0.52,
"learning_rate": 0.0005,
"loss": 2.0934,
"step": 147500
},
{
"epoch": 0.53,
"learning_rate": 0.0005,
"loss": 2.1133,
"step": 148000
},
{
"epoch": 0.53,
"learning_rate": 0.0005,
"loss": 2.0671,
"step": 148500
},
{
"epoch": 0.53,
"learning_rate": 0.0005,
"loss": 2.0993,
"step": 149000
},
{
"epoch": 0.53,
"learning_rate": 0.0005,
"loss": 2.1436,
"step": 149500
},
{
"epoch": 0.53,
"learning_rate": 0.0005,
"loss": 2.0909,
"step": 150000
},
{
"epoch": 0.53,
"eval_bleu": 13.1593,
"eval_gen_len": 17.2983,
"eval_runtime": 212.1864,
"eval_samples_per_second": 14.139,
"eval_steps_per_second": 1.767,
"step": 150000
},
{
"epoch": 0.54,
"learning_rate": 0.0005,
"loss": 2.0918,
"step": 150500
},
{
"epoch": 0.54,
"learning_rate": 0.0005,
"loss": 2.1168,
"step": 151000
},
{
"epoch": 0.54,
"learning_rate": 0.0005,
"loss": 2.0985,
"step": 151500
},
{
"epoch": 0.54,
"learning_rate": 0.0005,
"loss": 2.0956,
"step": 152000
},
{
"epoch": 0.54,
"learning_rate": 0.0005,
"loss": 2.1099,
"step": 152500
},
{
"epoch": 0.54,
"learning_rate": 0.0005,
"loss": 2.0837,
"step": 153000
},
{
"epoch": 0.55,
"learning_rate": 0.0005,
"loss": 2.098,
"step": 153500
},
{
"epoch": 0.55,
"learning_rate": 0.0005,
"loss": 2.1008,
"step": 154000
},
{
"epoch": 0.55,
"learning_rate": 0.0005,
"loss": 2.1101,
"step": 154500
},
{
"epoch": 0.55,
"learning_rate": 0.0005,
"loss": 2.0986,
"step": 155000
},
{
"epoch": 0.55,
"learning_rate": 0.0005,
"loss": 2.1101,
"step": 155500
},
{
"epoch": 0.55,
"learning_rate": 0.0005,
"loss": 2.0842,
"step": 156000
},
{
"epoch": 0.56,
"learning_rate": 0.0005,
"loss": 2.1082,
"step": 156500
},
{
"epoch": 0.56,
"learning_rate": 0.0005,
"loss": 2.0941,
"step": 157000
},
{
"epoch": 0.56,
"learning_rate": 0.0005,
"loss": 2.0751,
"step": 157500
},
{
"epoch": 0.56,
"learning_rate": 0.0005,
"loss": 2.0916,
"step": 158000
},
{
"epoch": 0.56,
"learning_rate": 0.0005,
"loss": 2.0802,
"step": 158500
},
{
"epoch": 0.57,
"learning_rate": 0.0005,
"loss": 2.0879,
"step": 159000
},
{
"epoch": 0.57,
"learning_rate": 0.0005,
"loss": 2.0932,
"step": 159500
},
{
"epoch": 0.57,
"learning_rate": 0.0005,
"loss": 2.0947,
"step": 160000
},
{
"epoch": 0.57,
"eval_bleu": 13.2815,
"eval_gen_len": 17.229,
"eval_runtime": 211.8324,
"eval_samples_per_second": 14.162,
"eval_steps_per_second": 1.77,
"step": 160000
},
{
"epoch": 0.57,
"learning_rate": 0.0005,
"loss": 2.0464,
"step": 160500
},
{
"epoch": 0.57,
"learning_rate": 0.0005,
"loss": 2.0811,
"step": 161000
},
{
"epoch": 0.57,
"learning_rate": 0.0005,
"loss": 2.0989,
"step": 161500
},
{
"epoch": 0.58,
"learning_rate": 0.0005,
"loss": 2.0888,
"step": 162000
},
{
"epoch": 0.58,
"learning_rate": 0.0005,
"loss": 2.0881,
"step": 162500
},
{
"epoch": 0.58,
"learning_rate": 0.0005,
"loss": 2.0728,
"step": 163000
},
{
"epoch": 0.58,
"learning_rate": 0.0005,
"loss": 2.0462,
"step": 163500
},
{
"epoch": 0.58,
"learning_rate": 0.0005,
"loss": 2.0762,
"step": 164000
},
{
"epoch": 0.59,
"learning_rate": 0.0005,
"loss": 2.0919,
"step": 164500
},
{
"epoch": 0.59,
"learning_rate": 0.0005,
"loss": 2.0736,
"step": 165000
},
{
"epoch": 0.59,
"learning_rate": 0.0005,
"loss": 2.0688,
"step": 165500
},
{
"epoch": 0.59,
"learning_rate": 0.0005,
"loss": 2.0949,
"step": 166000
},
{
"epoch": 0.59,
"learning_rate": 0.0005,
"loss": 2.0647,
"step": 166500
},
{
"epoch": 0.59,
"learning_rate": 0.0005,
"loss": 2.0942,
"step": 167000
},
{
"epoch": 0.6,
"learning_rate": 0.0005,
"loss": 2.0911,
"step": 167500
},
{
"epoch": 0.6,
"learning_rate": 0.0005,
"loss": 2.0855,
"step": 168000
},
{
"epoch": 0.6,
"learning_rate": 0.0005,
"loss": 2.0829,
"step": 168500
},
{
"epoch": 0.6,
"learning_rate": 0.0005,
"loss": 2.0626,
"step": 169000
},
{
"epoch": 0.6,
"learning_rate": 0.0005,
"loss": 2.0737,
"step": 169500
},
{
"epoch": 0.6,
"learning_rate": 0.0005,
"loss": 2.088,
"step": 170000
},
{
"epoch": 0.6,
"eval_bleu": 13.2294,
"eval_gen_len": 17.237,
"eval_runtime": 211.594,
"eval_samples_per_second": 14.178,
"eval_steps_per_second": 1.772,
"step": 170000
},
{
"epoch": 0.61,
"learning_rate": 0.0005,
"loss": 2.0452,
"step": 170500
},
{
"epoch": 0.61,
"learning_rate": 0.0005,
"loss": 2.0526,
"step": 171000
},
{
"epoch": 0.61,
"learning_rate": 0.0005,
"loss": 2.0784,
"step": 171500
},
{
"epoch": 0.61,
"learning_rate": 0.0005,
"loss": 2.0585,
"step": 172000
},
{
"epoch": 0.61,
"learning_rate": 0.0005,
"loss": 2.0615,
"step": 172500
},
{
"epoch": 0.62,
"learning_rate": 0.0005,
"loss": 2.0605,
"step": 173000
},
{
"epoch": 0.62,
"learning_rate": 0.0005,
"loss": 2.0624,
"step": 173500
},
{
"epoch": 0.62,
"learning_rate": 0.0005,
"loss": 2.0654,
"step": 174000
},
{
"epoch": 0.62,
"learning_rate": 0.0005,
"loss": 2.0464,
"step": 174500
},
{
"epoch": 0.62,
"learning_rate": 0.0005,
"loss": 2.0611,
"step": 175000
},
{
"epoch": 0.62,
"learning_rate": 0.0005,
"loss": 2.094,
"step": 175500
},
{
"epoch": 0.63,
"learning_rate": 0.0005,
"loss": 2.0565,
"step": 176000
},
{
"epoch": 0.63,
"learning_rate": 0.0005,
"loss": 2.057,
"step": 176500
},
{
"epoch": 0.63,
"learning_rate": 0.0005,
"loss": 2.0681,
"step": 177000
},
{
"epoch": 0.63,
"learning_rate": 0.0005,
"loss": 2.0316,
"step": 177500
},
{
"epoch": 0.63,
"learning_rate": 0.0005,
"loss": 2.0713,
"step": 178000
},
{
"epoch": 0.63,
"learning_rate": 0.0005,
"loss": 2.0786,
"step": 178500
},
{
"epoch": 0.64,
"learning_rate": 0.0005,
"loss": 2.0669,
"step": 179000
},
{
"epoch": 0.64,
"learning_rate": 0.0005,
"loss": 2.0505,
"step": 179500
},
{
"epoch": 0.64,
"learning_rate": 0.0005,
"loss": 2.0823,
"step": 180000
},
{
"epoch": 0.64,
"eval_bleu": 13.3117,
"eval_gen_len": 17.3163,
"eval_runtime": 210.849,
"eval_samples_per_second": 14.228,
"eval_steps_per_second": 1.779,
"step": 180000
},
{
"epoch": 0.64,
"learning_rate": 0.0005,
"loss": 2.0561,
"step": 180500
},
{
"epoch": 0.64,
"learning_rate": 0.0005,
"loss": 2.0381,
"step": 181000
},
{
"epoch": 0.65,
"learning_rate": 0.0005,
"loss": 2.0503,
"step": 181500
},
{
"epoch": 0.65,
"learning_rate": 0.0005,
"loss": 2.0706,
"step": 182000
},
{
"epoch": 0.65,
"learning_rate": 0.0005,
"loss": 2.0668,
"step": 182500
},
{
"epoch": 0.65,
"learning_rate": 0.0005,
"loss": 2.0657,
"step": 183000
},
{
"epoch": 0.65,
"learning_rate": 0.0005,
"loss": 2.0382,
"step": 183500
},
{
"epoch": 0.65,
"learning_rate": 0.0005,
"loss": 2.052,
"step": 184000
},
{
"epoch": 0.66,
"learning_rate": 0.0005,
"loss": 2.0755,
"step": 184500
},
{
"epoch": 0.66,
"learning_rate": 0.0005,
"loss": 2.0571,
"step": 185000
},
{
"epoch": 0.66,
"learning_rate": 0.0005,
"loss": 2.0192,
"step": 185500
},
{
"epoch": 0.66,
"learning_rate": 0.0005,
"loss": 2.0534,
"step": 186000
},
{
"epoch": 0.66,
"learning_rate": 0.0005,
"loss": 2.0545,
"step": 186500
},
{
"epoch": 0.67,
"learning_rate": 0.0005,
"loss": 2.0608,
"step": 187000
},
{
"epoch": 0.67,
"learning_rate": 0.0005,
"loss": 2.0693,
"step": 187500
},
{
"epoch": 0.67,
"learning_rate": 0.0005,
"loss": 2.0451,
"step": 188000
},
{
"epoch": 0.67,
"learning_rate": 0.0005,
"loss": 2.061,
"step": 188500
},
{
"epoch": 0.67,
"learning_rate": 0.0005,
"loss": 2.0575,
"step": 189000
},
{
"epoch": 0.67,
"learning_rate": 0.0005,
"loss": 2.0487,
"step": 189500
},
{
"epoch": 0.68,
"learning_rate": 0.0005,
"loss": 2.0386,
"step": 190000
},
{
"epoch": 0.68,
"eval_bleu": 13.3541,
"eval_gen_len": 17.222,
"eval_runtime": 211.7397,
"eval_samples_per_second": 14.168,
"eval_steps_per_second": 1.771,
"step": 190000
},
{
"epoch": 0.68,
"learning_rate": 0.0005,
"loss": 2.0442,
"step": 190500
},
{
"epoch": 0.68,
"learning_rate": 0.0005,
"loss": 2.0464,
"step": 191000
},
{
"epoch": 0.68,
"learning_rate": 0.0005,
"loss": 2.0499,
"step": 191500
},
{
"epoch": 0.68,
"learning_rate": 0.0005,
"loss": 2.0559,
"step": 192000
},
{
"epoch": 0.68,
"learning_rate": 0.0005,
"loss": 2.0488,
"step": 192500
},
{
"epoch": 0.69,
"learning_rate": 0.0005,
"loss": 2.0179,
"step": 193000
},
{
"epoch": 0.69,
"learning_rate": 0.0005,
"loss": 2.0409,
"step": 193500
},
{
"epoch": 0.69,
"learning_rate": 0.0005,
"loss": 2.0508,
"step": 194000
},
{
"epoch": 0.69,
"learning_rate": 0.0005,
"loss": 2.0173,
"step": 194500
},
{
"epoch": 0.69,
"learning_rate": 0.0005,
"loss": 2.0401,
"step": 195000
},
{
"epoch": 0.7,
"learning_rate": 0.0005,
"loss": 2.0539,
"step": 195500
},
{
"epoch": 0.7,
"learning_rate": 0.0005,
"loss": 2.0666,
"step": 196000
},
{
"epoch": 0.7,
"learning_rate": 0.0005,
"loss": 2.0347,
"step": 196500
},
{
"epoch": 0.7,
"learning_rate": 0.0005,
"loss": 2.0508,
"step": 197000
},
{
"epoch": 0.7,
"learning_rate": 0.0005,
"loss": 2.0324,
"step": 197500
},
{
"epoch": 0.7,
"learning_rate": 0.0005,
"loss": 2.0521,
"step": 198000
},
{
"epoch": 0.71,
"learning_rate": 0.0005,
"loss": 2.0399,
"step": 198500
},
{
"epoch": 0.71,
"learning_rate": 0.0005,
"loss": 2.052,
"step": 199000
},
{
"epoch": 0.71,
"learning_rate": 0.0005,
"loss": 2.0318,
"step": 199500
},
{
"epoch": 0.71,
"learning_rate": 0.0005,
"loss": 2.0431,
"step": 200000
},
{
"epoch": 0.71,
"eval_bleu": 13.6742,
"eval_gen_len": 17.2567,
"eval_runtime": 210.4095,
"eval_samples_per_second": 14.258,
"eval_steps_per_second": 1.782,
"step": 200000
},
{
"epoch": 0.71,
"learning_rate": 0.0005,
"loss": 2.0626,
"step": 200500
},
{
"epoch": 0.71,
"learning_rate": 0.0005,
"loss": 2.0585,
"step": 201000
},
{
"epoch": 0.72,
"learning_rate": 0.0005,
"loss": 2.0146,
"step": 201500
},
{
"epoch": 0.72,
"learning_rate": 0.0005,
"loss": 2.0478,
"step": 202000
},
{
"epoch": 0.72,
"learning_rate": 0.0005,
"loss": 2.0633,
"step": 202500
},
{
"epoch": 0.72,
"learning_rate": 0.0005,
"loss": 2.0407,
"step": 203000
},
{
"epoch": 0.72,
"learning_rate": 0.0005,
"loss": 2.0512,
"step": 203500
},
{
"epoch": 0.73,
"learning_rate": 0.0005,
"loss": 2.0373,
"step": 204000
},
{
"epoch": 0.73,
"learning_rate": 0.0005,
"loss": 2.0143,
"step": 204500
},
{
"epoch": 0.73,
"learning_rate": 0.0005,
"loss": 2.0394,
"step": 205000
},
{
"epoch": 0.73,
"learning_rate": 0.0005,
"loss": 2.0208,
"step": 205500
},
{
"epoch": 0.73,
"learning_rate": 0.0005,
"loss": 2.0145,
"step": 206000
},
{
"epoch": 0.73,
"learning_rate": 0.0005,
"loss": 2.0213,
"step": 206500
},
{
"epoch": 0.74,
"learning_rate": 0.0005,
"loss": 2.0238,
"step": 207000
},
{
"epoch": 0.74,
"learning_rate": 0.0005,
"loss": 2.0331,
"step": 207500
},
{
"epoch": 0.74,
"learning_rate": 0.0005,
"loss": 2.0456,
"step": 208000
},
{
"epoch": 0.74,
"learning_rate": 0.0005,
"loss": 2.0109,
"step": 208500
},
{
"epoch": 0.74,
"learning_rate": 0.0005,
"loss": 2.0352,
"step": 209000
},
{
"epoch": 0.75,
"learning_rate": 0.0005,
"loss": 2.0131,
"step": 209500
},
{
"epoch": 0.75,
"learning_rate": 0.0005,
"loss": 2.0419,
"step": 210000
},
{
"epoch": 0.75,
"eval_bleu": 13.2481,
"eval_gen_len": 17.2613,
"eval_runtime": 212.253,
"eval_samples_per_second": 14.134,
"eval_steps_per_second": 1.767,
"step": 210000
},
{
"epoch": 0.75,
"learning_rate": 0.0005,
"loss": 2.0275,
"step": 210500
},
{
"epoch": 0.75,
"learning_rate": 0.0005,
"loss": 2.0378,
"step": 211000
},
{
"epoch": 0.75,
"learning_rate": 0.0005,
"loss": 2.042,
"step": 211500
},
{
"epoch": 0.75,
"learning_rate": 0.0005,
"loss": 2.0247,
"step": 212000
},
{
"epoch": 0.76,
"learning_rate": 0.0005,
"loss": 2.0291,
"step": 212500
},
{
"epoch": 0.76,
"learning_rate": 0.0005,
"loss": 2.0191,
"step": 213000
},
{
"epoch": 0.76,
"learning_rate": 0.0005,
"loss": 2.0086,
"step": 213500
},
{
"epoch": 0.76,
"learning_rate": 0.0005,
"loss": 2.0508,
"step": 214000
},
{
"epoch": 0.76,
"learning_rate": 0.0005,
"loss": 2.0122,
"step": 214500
},
{
"epoch": 0.76,
"learning_rate": 0.0005,
"loss": 2.0123,
"step": 215000
},
{
"epoch": 0.77,
"learning_rate": 0.0005,
"loss": 2.0295,
"step": 215500
},
{
"epoch": 0.77,
"learning_rate": 0.0005,
"loss": 2.0152,
"step": 216000
},
{
"epoch": 0.77,
"learning_rate": 0.0005,
"loss": 2.0275,
"step": 216500
},
{
"epoch": 0.77,
"learning_rate": 0.0005,
"loss": 2.0317,
"step": 217000
},
{
"epoch": 0.77,
"learning_rate": 0.0005,
"loss": 1.9995,
"step": 217500
},
{
"epoch": 0.78,
"learning_rate": 0.0005,
"loss": 2.0266,
"step": 218000
},
{
"epoch": 0.78,
"learning_rate": 0.0005,
"loss": 2.0337,
"step": 218500
},
{
"epoch": 0.78,
"learning_rate": 0.0005,
"loss": 2.0399,
"step": 219000
},
{
"epoch": 0.78,
"learning_rate": 0.0005,
"loss": 2.0263,
"step": 219500
},
{
"epoch": 0.78,
"learning_rate": 0.0005,
"loss": 2.0097,
"step": 220000
},
{
"epoch": 0.78,
"eval_bleu": 13.6269,
"eval_gen_len": 17.29,
"eval_runtime": 210.1335,
"eval_samples_per_second": 14.277,
"eval_steps_per_second": 1.785,
"step": 220000
},
{
"epoch": 0.78,
"learning_rate": 0.0005,
"loss": 1.9963,
"step": 220500
},
{
"epoch": 0.79,
"learning_rate": 0.0005,
"loss": 2.0173,
"step": 221000
},
{
"epoch": 0.79,
"learning_rate": 0.0005,
"loss": 2.0184,
"step": 221500
},
{
"epoch": 0.79,
"learning_rate": 0.0005,
"loss": 1.99,
"step": 222000
},
{
"epoch": 0.79,
"learning_rate": 0.0005,
"loss": 2.0159,
"step": 222500
},
{
"epoch": 0.79,
"learning_rate": 0.0005,
"loss": 2.015,
"step": 223000
},
{
"epoch": 0.79,
"learning_rate": 0.0005,
"loss": 2.0349,
"step": 223500
},
{
"epoch": 0.8,
"learning_rate": 0.0005,
"loss": 2.015,
"step": 224000
},
{
"epoch": 0.8,
"learning_rate": 0.0005,
"loss": 2.0145,
"step": 224500
},
{
"epoch": 0.8,
"learning_rate": 0.0005,
"loss": 2.0142,
"step": 225000
},
{
"epoch": 0.8,
"learning_rate": 0.0005,
"loss": 2.0085,
"step": 225500
},
{
"epoch": 0.8,
"learning_rate": 0.0005,
"loss": 2.0099,
"step": 226000
},
{
"epoch": 0.81,
"learning_rate": 0.0005,
"loss": 2.0267,
"step": 226500
},
{
"epoch": 0.81,
"learning_rate": 0.0005,
"loss": 2.0294,
"step": 227000
},
{
"epoch": 0.81,
"learning_rate": 0.0005,
"loss": 2.0256,
"step": 227500
},
{
"epoch": 0.81,
"learning_rate": 0.0005,
"loss": 1.9883,
"step": 228000
},
{
"epoch": 0.81,
"learning_rate": 0.0005,
"loss": 2.0429,
"step": 228500
},
{
"epoch": 0.81,
"learning_rate": 0.0005,
"loss": 1.9926,
"step": 229000
},
{
"epoch": 0.82,
"learning_rate": 0.0005,
"loss": 2.0223,
"step": 229500
},
{
"epoch": 0.82,
"learning_rate": 0.0005,
"loss": 2.0184,
"step": 230000
},
{
"epoch": 0.82,
"eval_bleu": 13.5436,
"eval_gen_len": 17.262,
"eval_runtime": 211.2793,
"eval_samples_per_second": 14.199,
"eval_steps_per_second": 1.775,
"step": 230000
},
{
"epoch": 0.82,
"learning_rate": 0.0005,
"loss": 2.0164,
"step": 230500
},
{
"epoch": 0.82,
"learning_rate": 0.0005,
"loss": 2.01,
"step": 231000
},
{
"epoch": 0.82,
"learning_rate": 0.0005,
"loss": 2.0199,
"step": 231500
},
{
"epoch": 0.83,
"learning_rate": 0.0005,
"loss": 2.0162,
"step": 232000
},
{
"epoch": 0.83,
"learning_rate": 0.0005,
"loss": 1.9946,
"step": 232500
},
{
"epoch": 0.83,
"learning_rate": 0.0005,
"loss": 2.0271,
"step": 233000
},
{
"epoch": 0.83,
"learning_rate": 0.0005,
"loss": 2.0181,
"step": 233500
},
{
"epoch": 0.83,
"learning_rate": 0.0005,
"loss": 1.9774,
"step": 234000
},
{
"epoch": 0.83,
"learning_rate": 0.0005,
"loss": 1.9995,
"step": 234500
},
{
"epoch": 0.84,
"learning_rate": 0.0005,
"loss": 2.0043,
"step": 235000
},
{
"epoch": 0.84,
"learning_rate": 0.0005,
"loss": 1.9856,
"step": 235500
},
{
"epoch": 0.84,
"learning_rate": 0.0005,
"loss": 2.019,
"step": 236000
},
{
"epoch": 0.84,
"learning_rate": 0.0005,
"loss": 2.0151,
"step": 236500
},
{
"epoch": 0.84,
"learning_rate": 0.0005,
"loss": 1.9811,
"step": 237000
},
{
"epoch": 0.84,
"learning_rate": 0.0005,
"loss": 1.976,
"step": 237500
},
{
"epoch": 0.85,
"learning_rate": 0.0005,
"loss": 2.0171,
"step": 238000
},
{
"epoch": 0.85,
"learning_rate": 0.0005,
"loss": 2.0226,
"step": 238500
},
{
"epoch": 0.85,
"learning_rate": 0.0005,
"loss": 1.9925,
"step": 239000
},
{
"epoch": 0.85,
"learning_rate": 0.0005,
"loss": 2.0103,
"step": 239500
},
{
"epoch": 0.85,
"learning_rate": 0.0005,
"loss": 2.0321,
"step": 240000
},
{
"epoch": 0.85,
"eval_bleu": 13.691,
"eval_gen_len": 17.2677,
"eval_runtime": 211.7437,
"eval_samples_per_second": 14.168,
"eval_steps_per_second": 1.771,
"step": 240000
},
{
"epoch": 0.86,
"learning_rate": 0.0005,
"loss": 1.9975,
"step": 240500
},
{
"epoch": 0.86,
"learning_rate": 0.0005,
"loss": 1.9957,
"step": 241000
},
{
"epoch": 0.86,
"learning_rate": 0.0005,
"loss": 2.0085,
"step": 241500
},
{
"epoch": 0.86,
"learning_rate": 0.0005,
"loss": 2.0071,
"step": 242000
},
{
"epoch": 0.86,
"learning_rate": 0.0005,
"loss": 2.0153,
"step": 242500
},
{
"epoch": 0.86,
"learning_rate": 0.0005,
"loss": 2.0103,
"step": 243000
},
{
"epoch": 0.87,
"learning_rate": 0.0005,
"loss": 2.0012,
"step": 243500
},
{
"epoch": 0.87,
"learning_rate": 0.0005,
"loss": 1.9858,
"step": 244000
},
{
"epoch": 0.87,
"learning_rate": 0.0005,
"loss": 1.9878,
"step": 244500
},
{
"epoch": 0.87,
"learning_rate": 0.0005,
"loss": 2.0115,
"step": 245000
},
{
"epoch": 0.87,
"learning_rate": 0.0005,
"loss": 2.0136,
"step": 245500
},
{
"epoch": 0.87,
"learning_rate": 0.0005,
"loss": 2.0159,
"step": 246000
},
{
"epoch": 0.88,
"learning_rate": 0.0005,
"loss": 1.9778,
"step": 246500
},
{
"epoch": 0.88,
"learning_rate": 0.0005,
"loss": 1.976,
"step": 247000
},
{
"epoch": 0.88,
"learning_rate": 0.0005,
"loss": 2.0054,
"step": 247500
},
{
"epoch": 0.88,
"learning_rate": 0.0005,
"loss": 2.0282,
"step": 248000
},
{
"epoch": 0.88,
"learning_rate": 0.0005,
"loss": 1.9811,
"step": 248500
},
{
"epoch": 0.89,
"learning_rate": 0.0005,
"loss": 1.9773,
"step": 249000
},
{
"epoch": 0.89,
"learning_rate": 0.0005,
"loss": 1.9856,
"step": 249500
},
{
"epoch": 0.89,
"learning_rate": 0.0005,
"loss": 1.9836,
"step": 250000
},
{
"epoch": 0.89,
"eval_bleu": 13.8262,
"eval_gen_len": 17.2433,
"eval_runtime": 209.5455,
"eval_samples_per_second": 14.317,
"eval_steps_per_second": 1.79,
"step": 250000
},
{
"epoch": 0.89,
"learning_rate": 0.0005,
"loss": 2.0082,
"step": 250500
},
{
"epoch": 0.89,
"learning_rate": 0.0005,
"loss": 2.0006,
"step": 251000
},
{
"epoch": 0.89,
"learning_rate": 0.0005,
"loss": 2.0035,
"step": 251500
},
{
"epoch": 0.9,
"learning_rate": 0.0005,
"loss": 1.9641,
"step": 252000
},
{
"epoch": 0.9,
"learning_rate": 0.0005,
"loss": 1.9869,
"step": 252500
},
{
"epoch": 0.9,
"learning_rate": 0.0005,
"loss": 1.9964,
"step": 253000
},
{
"epoch": 0.9,
"learning_rate": 0.0005,
"loss": 2.0015,
"step": 253500
},
{
"epoch": 0.9,
"learning_rate": 0.0005,
"loss": 2.0152,
"step": 254000
},
{
"epoch": 0.91,
"learning_rate": 0.0005,
"loss": 1.9973,
"step": 254500
},
{
"epoch": 0.91,
"learning_rate": 0.0005,
"loss": 1.9884,
"step": 255000
},
{
"epoch": 0.91,
"learning_rate": 0.0005,
"loss": 2.007,
"step": 255500
},
{
"epoch": 0.91,
"learning_rate": 0.0005,
"loss": 2.0142,
"step": 256000
},
{
"epoch": 0.91,
"learning_rate": 0.0005,
"loss": 2.0124,
"step": 256500
},
{
"epoch": 0.91,
"learning_rate": 0.0005,
"loss": 2.0111,
"step": 257000
},
{
"epoch": 0.92,
"learning_rate": 0.0005,
"loss": 1.979,
"step": 257500
},
{
"epoch": 0.92,
"learning_rate": 0.0005,
"loss": 1.9947,
"step": 258000
},
{
"epoch": 0.92,
"learning_rate": 0.0005,
"loss": 1.985,
"step": 258500
},
{
"epoch": 0.92,
"learning_rate": 0.0005,
"loss": 1.9802,
"step": 259000
},
{
"epoch": 0.92,
"learning_rate": 0.0005,
"loss": 2.0062,
"step": 259500
},
{
"epoch": 0.92,
"learning_rate": 0.0005,
"loss": 1.9921,
"step": 260000
},
{
"epoch": 0.92,
"eval_bleu": 13.6876,
"eval_gen_len": 17.254,
"eval_runtime": 210.6894,
"eval_samples_per_second": 14.239,
"eval_steps_per_second": 1.78,
"step": 260000
},
{
"epoch": 0.93,
"learning_rate": 0.0005,
"loss": 1.9817,
"step": 260500
},
{
"epoch": 0.93,
"learning_rate": 0.0005,
"loss": 1.9949,
"step": 261000
},
{
"epoch": 0.93,
"learning_rate": 0.0005,
"loss": 1.9769,
"step": 261500
},
{
"epoch": 0.93,
"learning_rate": 0.0005,
"loss": 1.9957,
"step": 262000
},
{
"epoch": 0.93,
"learning_rate": 0.0005,
"loss": 2.0072,
"step": 262500
},
{
"epoch": 0.94,
"learning_rate": 0.0005,
"loss": 1.9858,
"step": 263000
},
{
"epoch": 0.94,
"learning_rate": 0.0005,
"loss": 1.9857,
"step": 263500
},
{
"epoch": 0.94,
"learning_rate": 0.0005,
"loss": 1.9634,
"step": 264000
},
{
"epoch": 0.94,
"learning_rate": 0.0005,
"loss": 1.9949,
"step": 264500
},
{
"epoch": 0.94,
"learning_rate": 0.0005,
"loss": 1.9898,
"step": 265000
},
{
"epoch": 0.94,
"learning_rate": 0.0005,
"loss": 2.0065,
"step": 265500
},
{
"epoch": 0.95,
"learning_rate": 0.0005,
"loss": 1.9879,
"step": 266000
},
{
"epoch": 0.95,
"learning_rate": 0.0005,
"loss": 1.9811,
"step": 266500
},
{
"epoch": 0.95,
"learning_rate": 0.0005,
"loss": 1.9486,
"step": 267000
},
{
"epoch": 0.95,
"learning_rate": 0.0005,
"loss": 1.9941,
"step": 267500
},
{
"epoch": 0.95,
"learning_rate": 0.0005,
"loss": 1.9992,
"step": 268000
},
{
"epoch": 0.95,
"learning_rate": 0.0005,
"loss": 2.0114,
"step": 268500
},
{
"epoch": 0.96,
"learning_rate": 0.0005,
"loss": 1.9735,
"step": 269000
},
{
"epoch": 0.96,
"learning_rate": 0.0005,
"loss": 1.9851,
"step": 269500
},
{
"epoch": 0.96,
"learning_rate": 0.0005,
"loss": 1.9978,
"step": 270000
},
{
"epoch": 0.96,
"eval_bleu": 13.5384,
"eval_gen_len": 17.2433,
"eval_runtime": 210.4598,
"eval_samples_per_second": 14.255,
"eval_steps_per_second": 1.782,
"step": 270000
},
{
"epoch": 0.96,
"learning_rate": 0.0005,
"loss": 2.0171,
"step": 270500
},
{
"epoch": 0.96,
"learning_rate": 0.0005,
"loss": 1.9665,
"step": 271000
},
{
"epoch": 0.97,
"learning_rate": 0.0005,
"loss": 1.977,
"step": 271500
},
{
"epoch": 0.97,
"learning_rate": 0.0005,
"loss": 1.9822,
"step": 272000
},
{
"epoch": 0.97,
"learning_rate": 0.0005,
"loss": 1.994,
"step": 272500
},
{
"epoch": 0.97,
"learning_rate": 0.0005,
"loss": 1.9753,
"step": 273000
},
{
"epoch": 0.97,
"learning_rate": 0.0005,
"loss": 1.975,
"step": 273500
},
{
"epoch": 0.97,
"learning_rate": 0.0005,
"loss": 1.9895,
"step": 274000
},
{
"epoch": 0.98,
"learning_rate": 0.0005,
"loss": 1.9764,
"step": 274500
},
{
"epoch": 0.98,
"learning_rate": 0.0005,
"loss": 1.9882,
"step": 275000
},
{
"epoch": 0.98,
"learning_rate": 0.0005,
"loss": 1.9879,
"step": 275500
},
{
"epoch": 0.98,
"learning_rate": 0.0005,
"loss": 1.9878,
"step": 276000
},
{
"epoch": 0.98,
"learning_rate": 0.0005,
"loss": 1.9501,
"step": 276500
},
{
"epoch": 0.99,
"learning_rate": 0.0005,
"loss": 1.9841,
"step": 277000
},
{
"epoch": 0.99,
"learning_rate": 0.0005,
"loss": 1.9862,
"step": 277500
},
{
"epoch": 0.99,
"learning_rate": 0.0005,
"loss": 1.9768,
"step": 278000
},
{
"epoch": 0.99,
"learning_rate": 0.0005,
"loss": 1.9893,
"step": 278500
},
{
"epoch": 0.99,
"learning_rate": 0.0005,
"loss": 1.9812,
"step": 279000
},
{
"epoch": 0.99,
"learning_rate": 0.0005,
"loss": 1.9696,
"step": 279500
},
{
"epoch": 1.0,
"learning_rate": 0.0005,
"loss": 1.9842,
"step": 280000
},
{
"epoch": 1.0,
"eval_bleu": 13.7593,
"eval_gen_len": 17.2593,
"eval_runtime": 208.654,
"eval_samples_per_second": 14.378,
"eval_steps_per_second": 1.797,
"step": 280000
},
{
"epoch": 1.0,
"learning_rate": 0.0005,
"loss": 1.979,
"step": 280500
},
{
"epoch": 1.0,
"learning_rate": 0.0005,
"loss": 1.9768,
"step": 281000
},
{
"epoch": 1.0,
"learning_rate": 0.0005,
"loss": 1.9605,
"step": 281500
},
{
"epoch": 1.0,
"learning_rate": 0.0005,
"loss": 1.9358,
"step": 282000
},
{
"epoch": 1.0,
"learning_rate": 0.0005,
"loss": 1.9724,
"step": 282500
},
{
"epoch": 1.01,
"learning_rate": 0.0005,
"loss": 1.935,
"step": 283000
},
{
"epoch": 1.01,
"learning_rate": 0.0005,
"loss": 1.93,
"step": 283500
},
{
"epoch": 1.01,
"learning_rate": 0.0005,
"loss": 1.9566,
"step": 284000
},
{
"epoch": 1.01,
"learning_rate": 0.0005,
"loss": 1.9272,
"step": 284500
},
{
"epoch": 1.01,
"learning_rate": 0.0005,
"loss": 1.9274,
"step": 285000
},
{
"epoch": 1.02,
"learning_rate": 0.0005,
"loss": 1.965,
"step": 285500
},
{
"epoch": 1.02,
"learning_rate": 0.0005,
"loss": 1.942,
"step": 286000
},
{
"epoch": 1.02,
"learning_rate": 0.0005,
"loss": 1.9553,
"step": 286500
},
{
"epoch": 1.02,
"learning_rate": 0.0005,
"loss": 1.9291,
"step": 287000
},
{
"epoch": 1.02,
"learning_rate": 0.0005,
"loss": 1.9528,
"step": 287500
},
{
"epoch": 1.02,
"learning_rate": 0.0005,
"loss": 1.9503,
"step": 288000
},
{
"epoch": 1.03,
"learning_rate": 0.0005,
"loss": 1.9189,
"step": 288500
},
{
"epoch": 1.03,
"learning_rate": 0.0005,
"loss": 1.9322,
"step": 289000
},
{
"epoch": 1.03,
"learning_rate": 0.0005,
"loss": 1.9466,
"step": 289500
},
{
"epoch": 1.03,
"learning_rate": 0.0005,
"loss": 1.9481,
"step": 290000
},
{
"epoch": 1.03,
"eval_bleu": 13.7266,
"eval_gen_len": 17.2537,
"eval_runtime": 210.3298,
"eval_samples_per_second": 14.263,
"eval_steps_per_second": 1.783,
"step": 290000
},
{
"epoch": 1.03,
"learning_rate": 0.0005,
"loss": 1.9061,
"step": 290500
},
{
"epoch": 1.03,
"learning_rate": 0.0005,
"loss": 1.9063,
"step": 291000
},
{
"epoch": 1.04,
"learning_rate": 0.0005,
"loss": 1.9396,
"step": 291500
},
{
"epoch": 1.04,
"learning_rate": 0.0005,
"loss": 1.9513,
"step": 292000
},
{
"epoch": 1.04,
"learning_rate": 0.0005,
"loss": 1.9753,
"step": 292500
},
{
"epoch": 1.04,
"learning_rate": 0.0005,
"loss": 1.9343,
"step": 293000
},
{
"epoch": 1.04,
"learning_rate": 0.0005,
"loss": 1.9548,
"step": 293500
},
{
"epoch": 1.05,
"learning_rate": 0.0005,
"loss": 1.9354,
"step": 294000
},
{
"epoch": 1.05,
"learning_rate": 0.0005,
"loss": 1.9324,
"step": 294500
},
{
"epoch": 1.05,
"learning_rate": 0.0005,
"loss": 1.9457,
"step": 295000
},
{
"epoch": 1.05,
"learning_rate": 0.0005,
"loss": 1.9278,
"step": 295500
},
{
"epoch": 1.05,
"learning_rate": 0.0005,
"loss": 1.901,
"step": 296000
},
{
"epoch": 1.05,
"learning_rate": 0.0005,
"loss": 1.9599,
"step": 296500
},
{
"epoch": 1.06,
"learning_rate": 0.0005,
"loss": 1.9593,
"step": 297000
},
{
"epoch": 1.06,
"learning_rate": 0.0005,
"loss": 1.9276,
"step": 297500
},
{
"epoch": 1.06,
"learning_rate": 0.0005,
"loss": 1.9408,
"step": 298000
},
{
"epoch": 1.06,
"learning_rate": 0.0005,
"loss": 1.9509,
"step": 298500
},
{
"epoch": 1.06,
"learning_rate": 0.0005,
"loss": 1.9509,
"step": 299000
},
{
"epoch": 1.07,
"learning_rate": 0.0005,
"loss": 1.9437,
"step": 299500
},
{
"epoch": 1.07,
"learning_rate": 0.0005,
"loss": 1.9808,
"step": 300000
},
{
"epoch": 1.07,
"eval_bleu": 13.9025,
"eval_gen_len": 17.217,
"eval_runtime": 210.0366,
"eval_samples_per_second": 14.283,
"eval_steps_per_second": 1.785,
"step": 300000
},
{
"epoch": 1.07,
"learning_rate": 0.0005,
"loss": 1.9484,
"step": 300500
},
{
"epoch": 1.07,
"learning_rate": 0.0005,
"loss": 1.9574,
"step": 301000
},
{
"epoch": 1.07,
"learning_rate": 0.0005,
"loss": 1.9477,
"step": 301500
},
{
"epoch": 1.07,
"learning_rate": 0.0005,
"loss": 1.9743,
"step": 302000
},
{
"epoch": 1.08,
"learning_rate": 0.0005,
"loss": 1.93,
"step": 302500
},
{
"epoch": 1.08,
"learning_rate": 0.0005,
"loss": 1.9117,
"step": 303000
},
{
"epoch": 1.08,
"learning_rate": 0.0005,
"loss": 1.9505,
"step": 303500
},
{
"epoch": 1.08,
"learning_rate": 0.0005,
"loss": 1.9717,
"step": 304000
},
{
"epoch": 1.08,
"learning_rate": 0.0005,
"loss": 1.9508,
"step": 304500
},
{
"epoch": 1.08,
"learning_rate": 0.0005,
"loss": 1.9408,
"step": 305000
},
{
"epoch": 1.09,
"learning_rate": 0.0005,
"loss": 1.9244,
"step": 305500
},
{
"epoch": 1.09,
"learning_rate": 0.0005,
"loss": 1.9406,
"step": 306000
},
{
"epoch": 1.09,
"learning_rate": 0.0005,
"loss": 1.9396,
"step": 306500
},
{
"epoch": 1.09,
"learning_rate": 0.0005,
"loss": 1.9509,
"step": 307000
},
{
"epoch": 1.09,
"learning_rate": 0.0005,
"loss": 1.9056,
"step": 307500
},
{
"epoch": 1.1,
"learning_rate": 0.0005,
"loss": 1.931,
"step": 308000
},
{
"epoch": 1.1,
"learning_rate": 0.0005,
"loss": 1.9528,
"step": 308500
},
{
"epoch": 1.1,
"learning_rate": 0.0005,
"loss": 1.9154,
"step": 309000
},
{
"epoch": 1.1,
"learning_rate": 0.0005,
"loss": 1.93,
"step": 309500
},
{
"epoch": 1.1,
"learning_rate": 0.0005,
"loss": 1.929,
"step": 310000
},
{
"epoch": 1.1,
"eval_bleu": 13.7188,
"eval_gen_len": 17.2703,
"eval_runtime": 209.7351,
"eval_samples_per_second": 14.304,
"eval_steps_per_second": 1.788,
"step": 310000
},
{
"epoch": 1.1,
"learning_rate": 0.0005,
"loss": 1.95,
"step": 310500
},
{
"epoch": 1.11,
"learning_rate": 0.0005,
"loss": 1.9401,
"step": 311000
},
{
"epoch": 1.11,
"learning_rate": 0.0005,
"loss": 1.943,
"step": 311500
},
{
"epoch": 1.11,
"learning_rate": 0.0005,
"loss": 1.9197,
"step": 312000
},
{
"epoch": 1.11,
"learning_rate": 0.0005,
"loss": 1.9421,
"step": 312500
},
{
"epoch": 1.11,
"learning_rate": 0.0005,
"loss": 1.9385,
"step": 313000
},
{
"epoch": 1.11,
"learning_rate": 0.0005,
"loss": 1.9182,
"step": 313500
},
{
"epoch": 1.12,
"learning_rate": 0.0005,
"loss": 1.9445,
"step": 314000
},
{
"epoch": 1.12,
"learning_rate": 0.0005,
"loss": 1.9132,
"step": 314500
},
{
"epoch": 1.12,
"learning_rate": 0.0005,
"loss": 1.9134,
"step": 315000
},
{
"epoch": 1.12,
"learning_rate": 0.0005,
"loss": 1.9551,
"step": 315500
},
{
"epoch": 1.12,
"learning_rate": 0.0005,
"loss": 1.9572,
"step": 316000
},
{
"epoch": 1.13,
"learning_rate": 0.0005,
"loss": 1.9672,
"step": 316500
},
{
"epoch": 1.13,
"learning_rate": 0.0005,
"loss": 1.9485,
"step": 317000
},
{
"epoch": 1.13,
"learning_rate": 0.0005,
"loss": 1.9575,
"step": 317500
},
{
"epoch": 1.13,
"learning_rate": 0.0005,
"loss": 1.9296,
"step": 318000
},
{
"epoch": 1.13,
"learning_rate": 0.0005,
"loss": 1.9572,
"step": 318500
},
{
"epoch": 1.13,
"learning_rate": 0.0005,
"loss": 1.9276,
"step": 319000
},
{
"epoch": 1.14,
"learning_rate": 0.0005,
"loss": 1.9512,
"step": 319500
},
{
"epoch": 1.14,
"learning_rate": 0.0005,
"loss": 1.9493,
"step": 320000
},
{
"epoch": 1.14,
"eval_bleu": 13.8795,
"eval_gen_len": 17.252,
"eval_runtime": 210.6151,
"eval_samples_per_second": 14.244,
"eval_steps_per_second": 1.78,
"step": 320000
},
{
"epoch": 1.14,
"learning_rate": 0.0005,
"loss": 1.9136,
"step": 320500
},
{
"epoch": 1.14,
"learning_rate": 0.0005,
"loss": 1.9206,
"step": 321000
},
{
"epoch": 1.14,
"learning_rate": 0.0005,
"loss": 1.9541,
"step": 321500
},
{
"epoch": 1.15,
"learning_rate": 0.0005,
"loss": 1.9498,
"step": 322000
},
{
"epoch": 1.15,
"learning_rate": 0.0005,
"loss": 1.9422,
"step": 322500
},
{
"epoch": 1.15,
"learning_rate": 0.0005,
"loss": 1.9356,
"step": 323000
},
{
"epoch": 1.15,
"learning_rate": 0.0005,
"loss": 1.9289,
"step": 323500
},
{
"epoch": 1.15,
"learning_rate": 0.0005,
"loss": 1.9355,
"step": 324000
},
{
"epoch": 1.15,
"learning_rate": 0.0005,
"loss": 1.9501,
"step": 324500
},
{
"epoch": 1.16,
"learning_rate": 0.0005,
"loss": 1.9092,
"step": 325000
},
{
"epoch": 1.16,
"learning_rate": 0.0005,
"loss": 1.9402,
"step": 325500
},
{
"epoch": 1.16,
"learning_rate": 0.0005,
"loss": 1.9522,
"step": 326000
},
{
"epoch": 1.16,
"learning_rate": 0.0005,
"loss": 1.9143,
"step": 326500
},
{
"epoch": 1.16,
"learning_rate": 0.0005,
"loss": 1.9259,
"step": 327000
},
{
"epoch": 1.16,
"learning_rate": 0.0005,
"loss": 1.9324,
"step": 327500
},
{
"epoch": 1.17,
"learning_rate": 0.0005,
"loss": 1.9349,
"step": 328000
},
{
"epoch": 1.17,
"learning_rate": 0.0005,
"loss": 1.9538,
"step": 328500
},
{
"epoch": 1.17,
"learning_rate": 0.0005,
"loss": 1.9365,
"step": 329000
},
{
"epoch": 1.17,
"learning_rate": 0.0005,
"loss": 1.9075,
"step": 329500
},
{
"epoch": 1.17,
"learning_rate": 0.0005,
"loss": 1.9176,
"step": 330000
},
{
"epoch": 1.17,
"eval_bleu": 13.8571,
"eval_gen_len": 17.2563,
"eval_runtime": 210.0443,
"eval_samples_per_second": 14.283,
"eval_steps_per_second": 1.785,
"step": 330000
},
{
"epoch": 1.18,
"learning_rate": 0.0005,
"loss": 1.9322,
"step": 330500
},
{
"epoch": 1.18,
"learning_rate": 0.0005,
"loss": 1.9333,
"step": 331000
},
{
"epoch": 1.18,
"learning_rate": 0.0005,
"loss": 1.9324,
"step": 331500
},
{
"epoch": 1.18,
"learning_rate": 0.0005,
"loss": 1.93,
"step": 332000
},
{
"epoch": 1.18,
"learning_rate": 0.0005,
"loss": 1.8769,
"step": 332500
},
{
"epoch": 1.18,
"learning_rate": 0.0005,
"loss": 1.9198,
"step": 333000
},
{
"epoch": 1.19,
"learning_rate": 0.0005,
"loss": 1.9084,
"step": 333500
},
{
"epoch": 1.19,
"learning_rate": 0.0005,
"loss": 1.9196,
"step": 334000
},
{
"epoch": 1.19,
"learning_rate": 0.0005,
"loss": 1.9354,
"step": 334500
},
{
"epoch": 1.19,
"learning_rate": 0.0005,
"loss": 1.9473,
"step": 335000
},
{
"epoch": 1.19,
"learning_rate": 0.0005,
"loss": 1.9315,
"step": 335500
},
{
"epoch": 1.19,
"learning_rate": 0.0005,
"loss": 1.9479,
"step": 336000
},
{
"epoch": 1.2,
"learning_rate": 0.0005,
"loss": 1.9078,
"step": 336500
},
{
"epoch": 1.2,
"learning_rate": 0.0005,
"loss": 1.9488,
"step": 337000
},
{
"epoch": 1.2,
"learning_rate": 0.0005,
"loss": 1.9399,
"step": 337500
},
{
"epoch": 1.2,
"learning_rate": 0.0005,
"loss": 1.9284,
"step": 338000
},
{
"epoch": 1.2,
"learning_rate": 0.0005,
"loss": 1.9436,
"step": 338500
},
{
"epoch": 1.21,
"learning_rate": 0.0005,
"loss": 1.9619,
"step": 339000
},
{
"epoch": 1.21,
"learning_rate": 0.0005,
"loss": 1.9424,
"step": 339500
},
{
"epoch": 1.21,
"learning_rate": 0.0005,
"loss": 1.933,
"step": 340000
},
{
"epoch": 1.21,
"eval_bleu": 13.951,
"eval_gen_len": 17.2883,
"eval_runtime": 210.4415,
"eval_samples_per_second": 14.256,
"eval_steps_per_second": 1.782,
"step": 340000
},
{
"epoch": 1.21,
"learning_rate": 0.0005,
"loss": 1.9235,
"step": 340500
},
{
"epoch": 1.21,
"learning_rate": 0.0005,
"loss": 1.9515,
"step": 341000
},
{
"epoch": 1.21,
"learning_rate": 0.0005,
"loss": 1.946,
"step": 341500
},
{
"epoch": 1.22,
"learning_rate": 0.0005,
"loss": 1.9255,
"step": 342000
},
{
"epoch": 1.22,
"learning_rate": 0.0005,
"loss": 1.9137,
"step": 342500
},
{
"epoch": 1.22,
"learning_rate": 0.0005,
"loss": 1.9264,
"step": 343000
},
{
"epoch": 1.22,
"learning_rate": 0.0005,
"loss": 1.9196,
"step": 343500
},
{
"epoch": 1.22,
"learning_rate": 0.0005,
"loss": 1.9168,
"step": 344000
},
{
"epoch": 1.23,
"learning_rate": 0.0005,
"loss": 1.9151,
"step": 344500
},
{
"epoch": 1.23,
"learning_rate": 0.0005,
"loss": 1.9499,
"step": 345000
},
{
"epoch": 1.23,
"learning_rate": 0.0005,
"loss": 1.9131,
"step": 345500
},
{
"epoch": 1.23,
"learning_rate": 0.0005,
"loss": 1.9427,
"step": 346000
},
{
"epoch": 1.23,
"learning_rate": 0.0005,
"loss": 1.9444,
"step": 346500
},
{
"epoch": 1.23,
"learning_rate": 0.0005,
"loss": 1.93,
"step": 347000
},
{
"epoch": 1.24,
"learning_rate": 0.0005,
"loss": 1.9173,
"step": 347500
},
{
"epoch": 1.24,
"learning_rate": 0.0005,
"loss": 1.9276,
"step": 348000
},
{
"epoch": 1.24,
"learning_rate": 0.0005,
"loss": 1.926,
"step": 348500
},
{
"epoch": 1.24,
"learning_rate": 0.0005,
"loss": 1.9242,
"step": 349000
},
{
"epoch": 1.24,
"learning_rate": 0.0005,
"loss": 1.9495,
"step": 349500
},
{
"epoch": 1.24,
"learning_rate": 0.0005,
"loss": 1.9164,
"step": 350000
},
{
"epoch": 1.24,
"eval_bleu": 13.9097,
"eval_gen_len": 17.2873,
"eval_runtime": 209.6162,
"eval_samples_per_second": 14.312,
"eval_steps_per_second": 1.789,
"step": 350000
},
{
"epoch": 1.25,
"learning_rate": 0.0005,
"loss": 1.9098,
"step": 350500
},
{
"epoch": 1.25,
"learning_rate": 0.0005,
"loss": 1.913,
"step": 351000
},
{
"epoch": 1.25,
"learning_rate": 0.0005,
"loss": 1.8882,
"step": 351500
},
{
"epoch": 1.25,
"learning_rate": 0.0005,
"loss": 1.8997,
"step": 352000
},
{
"epoch": 1.25,
"learning_rate": 0.0005,
"loss": 1.9239,
"step": 352500
},
{
"epoch": 1.26,
"learning_rate": 0.0005,
"loss": 1.9193,
"step": 353000
},
{
"epoch": 1.26,
"learning_rate": 0.0005,
"loss": 1.9476,
"step": 353500
},
{
"epoch": 1.26,
"learning_rate": 0.0005,
"loss": 1.9254,
"step": 354000
},
{
"epoch": 1.26,
"learning_rate": 0.0005,
"loss": 1.9267,
"step": 354500
},
{
"epoch": 1.26,
"learning_rate": 0.0005,
"loss": 1.9152,
"step": 355000
},
{
"epoch": 1.26,
"learning_rate": 0.0005,
"loss": 1.9012,
"step": 355500
},
{
"epoch": 1.27,
"learning_rate": 0.0005,
"loss": 1.9133,
"step": 356000
},
{
"epoch": 1.27,
"learning_rate": 0.0005,
"loss": 1.9281,
"step": 356500
},
{
"epoch": 1.27,
"learning_rate": 0.0005,
"loss": 1.9086,
"step": 357000
},
{
"epoch": 1.27,
"learning_rate": 0.0005,
"loss": 1.927,
"step": 357500
},
{
"epoch": 1.27,
"learning_rate": 0.0005,
"loss": 1.9396,
"step": 358000
},
{
"epoch": 1.27,
"learning_rate": 0.0005,
"loss": 1.9091,
"step": 358500
},
{
"epoch": 1.28,
"learning_rate": 0.0005,
"loss": 1.8993,
"step": 359000
},
{
"epoch": 1.28,
"learning_rate": 0.0005,
"loss": 1.9232,
"step": 359500
},
{
"epoch": 1.28,
"learning_rate": 0.0005,
"loss": 1.9021,
"step": 360000
},
{
"epoch": 1.28,
"eval_bleu": 13.9218,
"eval_gen_len": 17.2557,
"eval_runtime": 209.7441,
"eval_samples_per_second": 14.303,
"eval_steps_per_second": 1.788,
"step": 360000
},
{
"epoch": 1.28,
"learning_rate": 0.0005,
"loss": 1.9423,
"step": 360500
},
{
"epoch": 1.28,
"learning_rate": 0.0005,
"loss": 1.9263,
"step": 361000
},
{
"epoch": 1.29,
"learning_rate": 0.0005,
"loss": 1.8766,
"step": 361500
},
{
"epoch": 1.29,
"learning_rate": 0.0005,
"loss": 1.8903,
"step": 362000
},
{
"epoch": 1.29,
"learning_rate": 0.0005,
"loss": 1.9004,
"step": 362500
},
{
"epoch": 1.29,
"learning_rate": 0.0005,
"loss": 1.9262,
"step": 363000
},
{
"epoch": 1.29,
"learning_rate": 0.0005,
"loss": 1.9132,
"step": 363500
},
{
"epoch": 1.29,
"learning_rate": 0.0005,
"loss": 1.9317,
"step": 364000
},
{
"epoch": 1.3,
"learning_rate": 0.0005,
"loss": 1.9104,
"step": 364500
},
{
"epoch": 1.3,
"learning_rate": 0.0005,
"loss": 1.9239,
"step": 365000
},
{
"epoch": 1.3,
"learning_rate": 0.0005,
"loss": 1.932,
"step": 365500
},
{
"epoch": 1.3,
"learning_rate": 0.0005,
"loss": 1.9195,
"step": 366000
},
{
"epoch": 1.3,
"learning_rate": 0.0005,
"loss": 1.9197,
"step": 366500
},
{
"epoch": 1.31,
"learning_rate": 0.0005,
"loss": 1.8899,
"step": 367000
},
{
"epoch": 1.31,
"learning_rate": 0.0005,
"loss": 1.906,
"step": 367500
},
{
"epoch": 1.31,
"learning_rate": 0.0005,
"loss": 1.9475,
"step": 368000
},
{
"epoch": 1.31,
"learning_rate": 0.0005,
"loss": 1.9302,
"step": 368500
},
{
"epoch": 1.31,
"learning_rate": 0.0005,
"loss": 1.9071,
"step": 369000
},
{
"epoch": 1.31,
"learning_rate": 0.0005,
"loss": 1.9249,
"step": 369500
},
{
"epoch": 1.32,
"learning_rate": 0.0005,
"loss": 1.9229,
"step": 370000
},
{
"epoch": 1.32,
"eval_bleu": 13.9626,
"eval_gen_len": 17.2193,
"eval_runtime": 210.3966,
"eval_samples_per_second": 14.259,
"eval_steps_per_second": 1.782,
"step": 370000
},
{
"epoch": 1.32,
"learning_rate": 0.0005,
"loss": 1.9169,
"step": 370500
},
{
"epoch": 1.32,
"learning_rate": 0.0005,
"loss": 1.9311,
"step": 371000
},
{
"epoch": 1.32,
"learning_rate": 0.0005,
"loss": 1.9314,
"step": 371500
},
{
"epoch": 1.32,
"learning_rate": 0.0005,
"loss": 1.8933,
"step": 372000
},
{
"epoch": 1.32,
"learning_rate": 0.0005,
"loss": 1.915,
"step": 372500
},
{
"epoch": 1.33,
"learning_rate": 0.0005,
"loss": 1.9196,
"step": 373000
},
{
"epoch": 1.33,
"learning_rate": 0.0005,
"loss": 1.9201,
"step": 373500
},
{
"epoch": 1.33,
"learning_rate": 0.0005,
"loss": 1.9214,
"step": 374000
},
{
"epoch": 1.33,
"learning_rate": 0.0005,
"loss": 1.9269,
"step": 374500
},
{
"epoch": 1.33,
"learning_rate": 0.0005,
"loss": 1.9359,
"step": 375000
},
{
"epoch": 1.34,
"learning_rate": 0.0005,
"loss": 1.903,
"step": 375500
},
{
"epoch": 1.34,
"learning_rate": 0.0005,
"loss": 1.9154,
"step": 376000
},
{
"epoch": 1.34,
"learning_rate": 0.0005,
"loss": 1.9057,
"step": 376500
},
{
"epoch": 1.34,
"learning_rate": 0.0005,
"loss": 1.9249,
"step": 377000
},
{
"epoch": 1.34,
"learning_rate": 0.0005,
"loss": 1.9091,
"step": 377500
},
{
"epoch": 1.34,
"learning_rate": 0.0005,
"loss": 1.8887,
"step": 378000
},
{
"epoch": 1.35,
"learning_rate": 0.0005,
"loss": 1.9062,
"step": 378500
},
{
"epoch": 1.35,
"learning_rate": 0.0005,
"loss": 1.9382,
"step": 379000
},
{
"epoch": 1.35,
"learning_rate": 0.0005,
"loss": 1.9201,
"step": 379500
},
{
"epoch": 1.35,
"learning_rate": 0.0005,
"loss": 1.9052,
"step": 380000
},
{
"epoch": 1.35,
"eval_bleu": 13.9355,
"eval_gen_len": 17.1907,
"eval_runtime": 210.2658,
"eval_samples_per_second": 14.268,
"eval_steps_per_second": 1.783,
"step": 380000
},
{
"epoch": 1.35,
"learning_rate": 0.0005,
"loss": 1.9128,
"step": 380500
},
{
"epoch": 1.35,
"learning_rate": 0.0005,
"loss": 1.8844,
"step": 381000
},
{
"epoch": 1.36,
"learning_rate": 0.0005,
"loss": 1.9122,
"step": 381500
},
{
"epoch": 1.36,
"learning_rate": 0.0005,
"loss": 1.8976,
"step": 382000
},
{
"epoch": 1.36,
"learning_rate": 0.0005,
"loss": 1.9251,
"step": 382500
},
{
"epoch": 1.36,
"learning_rate": 0.0005,
"loss": 1.9067,
"step": 383000
},
{
"epoch": 1.36,
"learning_rate": 0.0005,
"loss": 1.9106,
"step": 383500
},
{
"epoch": 1.37,
"learning_rate": 0.0005,
"loss": 1.9069,
"step": 384000
},
{
"epoch": 1.37,
"learning_rate": 0.0005,
"loss": 1.9251,
"step": 384500
},
{
"epoch": 1.37,
"learning_rate": 0.0005,
"loss": 1.9339,
"step": 385000
},
{
"epoch": 1.37,
"learning_rate": 0.0005,
"loss": 1.9444,
"step": 385500
},
{
"epoch": 1.37,
"learning_rate": 0.0005,
"loss": 1.9154,
"step": 386000
},
{
"epoch": 1.37,
"learning_rate": 0.0005,
"loss": 1.9127,
"step": 386500
},
{
"epoch": 1.38,
"learning_rate": 0.0005,
"loss": 1.9214,
"step": 387000
},
{
"epoch": 1.38,
"learning_rate": 0.0005,
"loss": 1.8855,
"step": 387500
},
{
"epoch": 1.38,
"learning_rate": 0.0005,
"loss": 1.927,
"step": 388000
},
{
"epoch": 1.38,
"learning_rate": 0.0005,
"loss": 1.9108,
"step": 388500
},
{
"epoch": 1.38,
"learning_rate": 0.0005,
"loss": 1.9182,
"step": 389000
},
{
"epoch": 1.39,
"learning_rate": 0.0005,
"loss": 1.9008,
"step": 389500
},
{
"epoch": 1.39,
"learning_rate": 0.0005,
"loss": 1.9159,
"step": 390000
},
{
"epoch": 1.39,
"eval_bleu": 14.1,
"eval_gen_len": 17.2573,
"eval_runtime": 210.1633,
"eval_samples_per_second": 14.275,
"eval_steps_per_second": 1.784,
"step": 390000
},
{
"epoch": 1.39,
"learning_rate": 0.0005,
"loss": 1.8991,
"step": 390500
},
{
"epoch": 1.39,
"learning_rate": 0.0005,
"loss": 1.9076,
"step": 391000
},
{
"epoch": 1.39,
"learning_rate": 0.0005,
"loss": 1.9183,
"step": 391500
},
{
"epoch": 1.39,
"learning_rate": 0.0005,
"loss": 1.9114,
"step": 392000
},
{
"epoch": 1.4,
"learning_rate": 0.0005,
"loss": 1.9254,
"step": 392500
},
{
"epoch": 1.4,
"learning_rate": 0.0005,
"loss": 1.901,
"step": 393000
},
{
"epoch": 1.4,
"learning_rate": 0.0005,
"loss": 1.8843,
"step": 393500
},
{
"epoch": 1.4,
"learning_rate": 0.0005,
"loss": 1.9133,
"step": 394000
},
{
"epoch": 1.4,
"learning_rate": 0.0005,
"loss": 1.9348,
"step": 394500
},
{
"epoch": 1.4,
"learning_rate": 0.0005,
"loss": 1.9238,
"step": 395000
},
{
"epoch": 1.41,
"learning_rate": 0.0005,
"loss": 1.9106,
"step": 395500
},
{
"epoch": 1.41,
"learning_rate": 0.0005,
"loss": 1.9159,
"step": 396000
},
{
"epoch": 1.41,
"learning_rate": 0.0005,
"loss": 1.9189,
"step": 396500
},
{
"epoch": 1.41,
"learning_rate": 0.0005,
"loss": 1.9114,
"step": 397000
},
{
"epoch": 1.41,
"learning_rate": 0.0005,
"loss": 1.904,
"step": 397500
},
{
"epoch": 1.42,
"learning_rate": 0.0005,
"loss": 1.8842,
"step": 398000
},
{
"epoch": 1.42,
"learning_rate": 0.0005,
"loss": 1.903,
"step": 398500
},
{
"epoch": 1.42,
"learning_rate": 0.0005,
"loss": 1.9161,
"step": 399000
},
{
"epoch": 1.42,
"learning_rate": 0.0005,
"loss": 1.939,
"step": 399500
},
{
"epoch": 1.42,
"learning_rate": 0.0005,
"loss": 1.8946,
"step": 400000
},
{
"epoch": 1.42,
"eval_bleu": 14.2056,
"eval_gen_len": 17.2387,
"eval_runtime": 209.8848,
"eval_samples_per_second": 14.294,
"eval_steps_per_second": 1.787,
"step": 400000
},
{
"epoch": 1.42,
"learning_rate": 0.0005,
"loss": 1.9221,
"step": 400500
},
{
"epoch": 1.43,
"learning_rate": 0.0005,
"loss": 1.9273,
"step": 401000
},
{
"epoch": 1.43,
"learning_rate": 0.0005,
"loss": 1.9062,
"step": 401500
},
{
"epoch": 1.43,
"learning_rate": 0.0005,
"loss": 1.8952,
"step": 402000
},
{
"epoch": 1.43,
"learning_rate": 0.0005,
"loss": 1.9129,
"step": 402500
},
{
"epoch": 1.43,
"learning_rate": 0.0005,
"loss": 1.9197,
"step": 403000
},
{
"epoch": 1.44,
"learning_rate": 0.0005,
"loss": 1.8995,
"step": 403500
},
{
"epoch": 1.44,
"learning_rate": 0.0005,
"loss": 1.9148,
"step": 404000
},
{
"epoch": 1.44,
"learning_rate": 0.0005,
"loss": 1.8738,
"step": 404500
},
{
"epoch": 1.44,
"learning_rate": 0.0005,
"loss": 1.9039,
"step": 405000
},
{
"epoch": 1.44,
"learning_rate": 0.0005,
"loss": 1.8796,
"step": 405500
},
{
"epoch": 1.44,
"learning_rate": 0.0005,
"loss": 1.8763,
"step": 406000
},
{
"epoch": 1.45,
"learning_rate": 0.0005,
"loss": 1.8879,
"step": 406500
},
{
"epoch": 1.45,
"learning_rate": 0.0005,
"loss": 1.9255,
"step": 407000
},
{
"epoch": 1.45,
"learning_rate": 0.0005,
"loss": 1.8904,
"step": 407500
},
{
"epoch": 1.45,
"learning_rate": 0.0005,
"loss": 1.9185,
"step": 408000
},
{
"epoch": 1.45,
"learning_rate": 0.0005,
"loss": 1.9233,
"step": 408500
},
{
"epoch": 1.45,
"learning_rate": 0.0005,
"loss": 1.9053,
"step": 409000
},
{
"epoch": 1.46,
"learning_rate": 0.0005,
"loss": 1.9257,
"step": 409500
},
{
"epoch": 1.46,
"learning_rate": 0.0005,
"loss": 1.8702,
"step": 410000
},
{
"epoch": 1.46,
"eval_bleu": 14.1317,
"eval_gen_len": 17.2767,
"eval_runtime": 209.9792,
"eval_samples_per_second": 14.287,
"eval_steps_per_second": 1.786,
"step": 410000
},
{
"epoch": 1.46,
"learning_rate": 0.0005,
"loss": 1.9006,
"step": 410500
},
{
"epoch": 1.46,
"learning_rate": 0.0005,
"loss": 1.9165,
"step": 411000
},
{
"epoch": 1.46,
"learning_rate": 0.0005,
"loss": 1.8969,
"step": 411500
},
{
"epoch": 1.47,
"learning_rate": 0.0005,
"loss": 1.9103,
"step": 412000
},
{
"epoch": 1.47,
"learning_rate": 0.0005,
"loss": 1.9102,
"step": 412500
},
{
"epoch": 1.47,
"learning_rate": 0.0005,
"loss": 1.8816,
"step": 413000
},
{
"epoch": 1.47,
"learning_rate": 0.0005,
"loss": 1.8814,
"step": 413500
},
{
"epoch": 1.47,
"learning_rate": 0.0005,
"loss": 1.9083,
"step": 414000
},
{
"epoch": 1.47,
"learning_rate": 0.0005,
"loss": 1.903,
"step": 414500
},
{
"epoch": 1.48,
"learning_rate": 0.0005,
"loss": 1.9214,
"step": 415000
},
{
"epoch": 1.48,
"learning_rate": 0.0005,
"loss": 1.8792,
"step": 415500
},
{
"epoch": 1.48,
"learning_rate": 0.0005,
"loss": 1.9242,
"step": 416000
},
{
"epoch": 1.48,
"learning_rate": 0.0005,
"loss": 1.8963,
"step": 416500
},
{
"epoch": 1.48,
"learning_rate": 0.0005,
"loss": 1.9194,
"step": 417000
},
{
"epoch": 1.48,
"learning_rate": 0.0005,
"loss": 1.9021,
"step": 417500
},
{
"epoch": 1.49,
"learning_rate": 0.0005,
"loss": 1.8832,
"step": 418000
},
{
"epoch": 1.49,
"learning_rate": 0.0005,
"loss": 1.8885,
"step": 418500
},
{
"epoch": 1.49,
"learning_rate": 0.0005,
"loss": 1.9143,
"step": 419000
},
{
"epoch": 1.49,
"learning_rate": 0.0005,
"loss": 1.8917,
"step": 419500
},
{
"epoch": 1.49,
"learning_rate": 0.0005,
"loss": 1.8957,
"step": 420000
},
{
"epoch": 1.49,
"eval_bleu": 13.8621,
"eval_gen_len": 17.2537,
"eval_runtime": 210.8259,
"eval_samples_per_second": 14.23,
"eval_steps_per_second": 1.779,
"step": 420000
},
{
"epoch": 1.5,
"learning_rate": 0.0005,
"loss": 1.9073,
"step": 420500
},
{
"epoch": 1.5,
"learning_rate": 0.0005,
"loss": 1.8913,
"step": 421000
},
{
"epoch": 1.5,
"learning_rate": 0.0005,
"loss": 1.8717,
"step": 421500
},
{
"epoch": 1.5,
"learning_rate": 0.0005,
"loss": 1.8888,
"step": 422000
},
{
"epoch": 1.5,
"learning_rate": 0.0005,
"loss": 1.8941,
"step": 422500
},
{
"epoch": 1.5,
"learning_rate": 0.0005,
"loss": 1.892,
"step": 423000
},
{
"epoch": 1.51,
"learning_rate": 0.0005,
"loss": 1.8733,
"step": 423500
},
{
"epoch": 1.51,
"learning_rate": 0.0005,
"loss": 1.909,
"step": 424000
},
{
"epoch": 1.51,
"learning_rate": 0.0005,
"loss": 1.8916,
"step": 424500
},
{
"epoch": 1.51,
"learning_rate": 0.0005,
"loss": 1.8855,
"step": 425000
},
{
"epoch": 1.51,
"learning_rate": 0.0005,
"loss": 1.9027,
"step": 425500
},
{
"epoch": 1.52,
"learning_rate": 0.0005,
"loss": 1.8949,
"step": 426000
},
{
"epoch": 1.52,
"learning_rate": 0.0005,
"loss": 1.9005,
"step": 426500
},
{
"epoch": 1.52,
"learning_rate": 0.0005,
"loss": 1.9224,
"step": 427000
},
{
"epoch": 1.52,
"learning_rate": 0.0005,
"loss": 1.886,
"step": 427500
},
{
"epoch": 1.52,
"learning_rate": 0.0005,
"loss": 1.8956,
"step": 428000
},
{
"epoch": 1.52,
"learning_rate": 0.0005,
"loss": 1.8748,
"step": 428500
},
{
"epoch": 1.53,
"learning_rate": 0.0005,
"loss": 1.8819,
"step": 429000
},
{
"epoch": 1.53,
"learning_rate": 0.0005,
"loss": 1.9058,
"step": 429500
},
{
"epoch": 1.53,
"learning_rate": 0.0005,
"loss": 1.8796,
"step": 430000
},
{
"epoch": 1.53,
"eval_bleu": 14.1927,
"eval_gen_len": 17.2637,
"eval_runtime": 209.2327,
"eval_samples_per_second": 14.338,
"eval_steps_per_second": 1.792,
"step": 430000
},
{
"epoch": 1.53,
"learning_rate": 0.0005,
"loss": 1.9126,
"step": 430500
},
{
"epoch": 1.53,
"learning_rate": 0.0005,
"loss": 1.9029,
"step": 431000
},
{
"epoch": 1.53,
"learning_rate": 0.0005,
"loss": 1.8999,
"step": 431500
},
{
"epoch": 1.54,
"learning_rate": 0.0005,
"loss": 1.9121,
"step": 432000
},
{
"epoch": 1.54,
"learning_rate": 0.0005,
"loss": 1.9198,
"step": 432500
},
{
"epoch": 1.54,
"learning_rate": 0.0005,
"loss": 1.8766,
"step": 433000
},
{
"epoch": 1.54,
"learning_rate": 0.0005,
"loss": 1.9225,
"step": 433500
},
{
"epoch": 1.54,
"learning_rate": 0.0005,
"loss": 1.92,
"step": 434000
},
{
"epoch": 1.55,
"learning_rate": 0.0005,
"loss": 1.9178,
"step": 434500
},
{
"epoch": 1.55,
"learning_rate": 0.0005,
"loss": 1.88,
"step": 435000
},
{
"epoch": 1.55,
"learning_rate": 0.0005,
"loss": 1.8936,
"step": 435500
},
{
"epoch": 1.55,
"learning_rate": 0.0005,
"loss": 1.9063,
"step": 436000
},
{
"epoch": 1.55,
"learning_rate": 0.0005,
"loss": 1.8763,
"step": 436500
},
{
"epoch": 1.55,
"learning_rate": 0.0005,
"loss": 1.9093,
"step": 437000
},
{
"epoch": 1.56,
"learning_rate": 0.0005,
"loss": 1.9274,
"step": 437500
},
{
"epoch": 1.56,
"learning_rate": 0.0005,
"loss": 1.8963,
"step": 438000
},
{
"epoch": 1.56,
"learning_rate": 0.0005,
"loss": 1.9107,
"step": 438500
},
{
"epoch": 1.56,
"learning_rate": 0.0005,
"loss": 1.9044,
"step": 439000
},
{
"epoch": 1.56,
"learning_rate": 0.0005,
"loss": 1.893,
"step": 439500
},
{
"epoch": 1.56,
"learning_rate": 0.0005,
"loss": 1.895,
"step": 440000
},
{
"epoch": 1.56,
"eval_bleu": 14.0582,
"eval_gen_len": 17.2333,
"eval_runtime": 208.5065,
"eval_samples_per_second": 14.388,
"eval_steps_per_second": 1.799,
"step": 440000
},
{
"epoch": 1.57,
"learning_rate": 0.0005,
"loss": 1.9065,
"step": 440500
},
{
"epoch": 1.57,
"learning_rate": 0.0005,
"loss": 1.8946,
"step": 441000
},
{
"epoch": 1.57,
"learning_rate": 0.0005,
"loss": 1.9015,
"step": 441500
},
{
"epoch": 1.57,
"learning_rate": 0.0005,
"loss": 1.8853,
"step": 442000
},
{
"epoch": 1.57,
"learning_rate": 0.0005,
"loss": 1.8862,
"step": 442500
},
{
"epoch": 1.58,
"learning_rate": 0.0005,
"loss": 1.8777,
"step": 443000
},
{
"epoch": 1.58,
"learning_rate": 0.0005,
"loss": 1.8956,
"step": 443500
},
{
"epoch": 1.58,
"learning_rate": 0.0005,
"loss": 1.8855,
"step": 444000
},
{
"epoch": 1.58,
"learning_rate": 0.0005,
"loss": 1.8992,
"step": 444500
},
{
"epoch": 1.58,
"learning_rate": 0.0005,
"loss": 1.8788,
"step": 445000
},
{
"epoch": 1.58,
"learning_rate": 0.0005,
"loss": 1.8861,
"step": 445500
},
{
"epoch": 1.59,
"learning_rate": 0.0005,
"loss": 1.9067,
"step": 446000
},
{
"epoch": 1.59,
"learning_rate": 0.0005,
"loss": 1.9006,
"step": 446500
},
{
"epoch": 1.59,
"learning_rate": 0.0005,
"loss": 1.9093,
"step": 447000
},
{
"epoch": 1.59,
"learning_rate": 0.0005,
"loss": 1.876,
"step": 447500
},
{
"epoch": 1.59,
"learning_rate": 0.0005,
"loss": 1.8703,
"step": 448000
},
{
"epoch": 1.6,
"learning_rate": 0.0005,
"loss": 1.86,
"step": 448500
},
{
"epoch": 1.6,
"learning_rate": 0.0005,
"loss": 1.8887,
"step": 449000
},
{
"epoch": 1.6,
"learning_rate": 0.0005,
"loss": 1.8746,
"step": 449500
},
{
"epoch": 1.6,
"learning_rate": 0.0005,
"loss": 1.8869,
"step": 450000
},
{
"epoch": 1.6,
"eval_bleu": 14.0195,
"eval_gen_len": 17.265,
"eval_runtime": 208.4219,
"eval_samples_per_second": 14.394,
"eval_steps_per_second": 1.799,
"step": 450000
},
{
"epoch": 1.6,
"learning_rate": 0.0005,
"loss": 1.9001,
"step": 450500
},
{
"epoch": 1.6,
"learning_rate": 0.0005,
"loss": 1.8942,
"step": 451000
},
{
"epoch": 1.61,
"learning_rate": 0.0005,
"loss": 1.8915,
"step": 451500
},
{
"epoch": 1.61,
"learning_rate": 0.0005,
"loss": 1.8738,
"step": 452000
},
{
"epoch": 1.61,
"learning_rate": 0.0005,
"loss": 1.894,
"step": 452500
},
{
"epoch": 1.61,
"learning_rate": 0.0005,
"loss": 1.9108,
"step": 453000
},
{
"epoch": 1.61,
"learning_rate": 0.0005,
"loss": 1.8666,
"step": 453500
},
{
"epoch": 1.61,
"learning_rate": 0.0005,
"loss": 1.8989,
"step": 454000
},
{
"epoch": 1.62,
"learning_rate": 0.0005,
"loss": 1.8975,
"step": 454500
},
{
"epoch": 1.62,
"learning_rate": 0.0005,
"loss": 1.8896,
"step": 455000
},
{
"epoch": 1.62,
"learning_rate": 0.0005,
"loss": 1.903,
"step": 455500
},
{
"epoch": 1.62,
"learning_rate": 0.0005,
"loss": 1.9147,
"step": 456000
},
{
"epoch": 1.62,
"learning_rate": 0.0005,
"loss": 1.8979,
"step": 456500
},
{
"epoch": 1.63,
"learning_rate": 0.0005,
"loss": 1.8686,
"step": 457000
},
{
"epoch": 1.63,
"learning_rate": 0.0005,
"loss": 1.9064,
"step": 457500
},
{
"epoch": 1.63,
"learning_rate": 0.0005,
"loss": 1.891,
"step": 458000
},
{
"epoch": 1.63,
"learning_rate": 0.0005,
"loss": 1.9,
"step": 458500
},
{
"epoch": 1.63,
"learning_rate": 0.0005,
"loss": 1.8841,
"step": 459000
},
{
"epoch": 1.63,
"learning_rate": 0.0005,
"loss": 1.8899,
"step": 459500
},
{
"epoch": 1.64,
"learning_rate": 0.0005,
"loss": 1.8935,
"step": 460000
},
{
"epoch": 1.64,
"eval_bleu": 14.0753,
"eval_gen_len": 17.2433,
"eval_runtime": 207.8742,
"eval_samples_per_second": 14.432,
"eval_steps_per_second": 1.804,
"step": 460000
},
{
"epoch": 1.64,
"learning_rate": 0.0005,
"loss": 1.8822,
"step": 460500
},
{
"epoch": 1.64,
"learning_rate": 0.0005,
"loss": 1.8702,
"step": 461000
},
{
"epoch": 1.64,
"learning_rate": 0.0005,
"loss": 1.8791,
"step": 461500
},
{
"epoch": 1.64,
"learning_rate": 0.0005,
"loss": 1.8787,
"step": 462000
},
{
"epoch": 1.64,
"learning_rate": 0.0005,
"loss": 1.8697,
"step": 462500
},
{
"epoch": 1.65,
"learning_rate": 0.0005,
"loss": 1.9087,
"step": 463000
},
{
"epoch": 1.65,
"learning_rate": 0.0005,
"loss": 1.8696,
"step": 463500
},
{
"epoch": 1.65,
"learning_rate": 0.0005,
"loss": 1.8831,
"step": 464000
},
{
"epoch": 1.65,
"learning_rate": 0.0005,
"loss": 1.913,
"step": 464500
},
{
"epoch": 1.65,
"learning_rate": 0.0005,
"loss": 1.8784,
"step": 465000
},
{
"epoch": 1.66,
"learning_rate": 0.0005,
"loss": 1.8686,
"step": 465500
},
{
"epoch": 1.66,
"learning_rate": 0.0005,
"loss": 1.8773,
"step": 466000
},
{
"epoch": 1.66,
"learning_rate": 0.0005,
"loss": 1.8849,
"step": 466500
},
{
"epoch": 1.66,
"learning_rate": 0.0005,
"loss": 1.8775,
"step": 467000
},
{
"epoch": 1.66,
"learning_rate": 0.0005,
"loss": 1.8864,
"step": 467500
},
{
"epoch": 1.66,
"learning_rate": 0.0005,
"loss": 1.9001,
"step": 468000
},
{
"epoch": 1.67,
"learning_rate": 0.0005,
"loss": 1.9055,
"step": 468500
},
{
"epoch": 1.67,
"learning_rate": 0.0005,
"loss": 1.8874,
"step": 469000
},
{
"epoch": 1.67,
"learning_rate": 0.0005,
"loss": 1.8834,
"step": 469500
},
{
"epoch": 1.67,
"learning_rate": 0.0005,
"loss": 1.8775,
"step": 470000
},
{
"epoch": 1.67,
"eval_bleu": 14.1,
"eval_gen_len": 17.2593,
"eval_runtime": 208.9726,
"eval_samples_per_second": 14.356,
"eval_steps_per_second": 1.794,
"step": 470000
},
{
"epoch": 1.67,
"learning_rate": 0.0005,
"loss": 1.8792,
"step": 470500
},
{
"epoch": 1.68,
"learning_rate": 0.0005,
"loss": 1.8794,
"step": 471000
},
{
"epoch": 1.68,
"learning_rate": 0.0005,
"loss": 1.8797,
"step": 471500
},
{
"epoch": 1.68,
"learning_rate": 0.0005,
"loss": 1.8709,
"step": 472000
},
{
"epoch": 1.68,
"learning_rate": 0.0005,
"loss": 1.9154,
"step": 472500
},
{
"epoch": 1.68,
"learning_rate": 0.0005,
"loss": 1.8894,
"step": 473000
},
{
"epoch": 1.68,
"learning_rate": 0.0005,
"loss": 1.8758,
"step": 473500
},
{
"epoch": 1.69,
"learning_rate": 0.0005,
"loss": 1.8954,
"step": 474000
},
{
"epoch": 1.69,
"learning_rate": 0.0005,
"loss": 1.8666,
"step": 474500
},
{
"epoch": 1.69,
"learning_rate": 0.0005,
"loss": 1.8768,
"step": 475000
},
{
"epoch": 1.69,
"learning_rate": 0.0005,
"loss": 1.8835,
"step": 475500
},
{
"epoch": 1.69,
"learning_rate": 0.0005,
"loss": 1.8768,
"step": 476000
},
{
"epoch": 1.69,
"learning_rate": 0.0005,
"loss": 1.8924,
"step": 476500
},
{
"epoch": 1.7,
"learning_rate": 0.0005,
"loss": 1.8866,
"step": 477000
},
{
"epoch": 1.7,
"learning_rate": 0.0005,
"loss": 1.8736,
"step": 477500
},
{
"epoch": 1.7,
"learning_rate": 0.0005,
"loss": 1.883,
"step": 478000
},
{
"epoch": 1.7,
"learning_rate": 0.0005,
"loss": 1.8977,
"step": 478500
},
{
"epoch": 1.7,
"learning_rate": 0.0005,
"loss": 1.8602,
"step": 479000
},
{
"epoch": 1.71,
"learning_rate": 0.0005,
"loss": 1.8827,
"step": 479500
},
{
"epoch": 1.71,
"learning_rate": 0.0005,
"loss": 1.8683,
"step": 480000
},
{
"epoch": 1.71,
"eval_bleu": 14.2598,
"eval_gen_len": 17.2647,
"eval_runtime": 209.7872,
"eval_samples_per_second": 14.3,
"eval_steps_per_second": 1.788,
"step": 480000
},
{
"epoch": 1.71,
"learning_rate": 0.0005,
"loss": 1.8754,
"step": 480500
},
{
"epoch": 1.71,
"learning_rate": 0.0005,
"loss": 1.8634,
"step": 481000
},
{
"epoch": 1.71,
"learning_rate": 0.0005,
"loss": 1.9076,
"step": 481500
},
{
"epoch": 1.71,
"learning_rate": 0.0005,
"loss": 1.8956,
"step": 482000
},
{
"epoch": 1.72,
"learning_rate": 0.0005,
"loss": 1.8705,
"step": 482500
},
{
"epoch": 1.72,
"learning_rate": 0.0005,
"loss": 1.8754,
"step": 483000
},
{
"epoch": 1.72,
"learning_rate": 0.0005,
"loss": 1.8641,
"step": 483500
},
{
"epoch": 1.72,
"learning_rate": 0.0005,
"loss": 1.8731,
"step": 484000
},
{
"epoch": 1.72,
"learning_rate": 0.0005,
"loss": 1.8836,
"step": 484500
},
{
"epoch": 1.72,
"learning_rate": 0.0005,
"loss": 1.8857,
"step": 485000
},
{
"epoch": 1.73,
"learning_rate": 0.0005,
"loss": 1.8813,
"step": 485500
},
{
"epoch": 1.73,
"learning_rate": 0.0005,
"loss": 1.8884,
"step": 486000
},
{
"epoch": 1.73,
"learning_rate": 0.0005,
"loss": 1.9007,
"step": 486500
},
{
"epoch": 1.73,
"learning_rate": 0.0005,
"loss": 1.8875,
"step": 487000
},
{
"epoch": 1.73,
"learning_rate": 0.0005,
"loss": 1.8734,
"step": 487500
},
{
"epoch": 1.74,
"learning_rate": 0.0005,
"loss": 1.8457,
"step": 488000
},
{
"epoch": 1.74,
"learning_rate": 0.0005,
"loss": 1.9094,
"step": 488500
},
{
"epoch": 1.74,
"learning_rate": 0.0005,
"loss": 1.8856,
"step": 489000
},
{
"epoch": 1.74,
"learning_rate": 0.0005,
"loss": 1.8534,
"step": 489500
},
{
"epoch": 1.74,
"learning_rate": 0.0005,
"loss": 1.8511,
"step": 490000
},
{
"epoch": 1.74,
"eval_bleu": 14.3516,
"eval_gen_len": 17.2557,
"eval_runtime": 208.2762,
"eval_samples_per_second": 14.404,
"eval_steps_per_second": 1.8,
"step": 490000
},
{
"epoch": 1.74,
"learning_rate": 0.0005,
"loss": 1.9023,
"step": 490500
},
{
"epoch": 1.75,
"learning_rate": 0.0005,
"loss": 1.8887,
"step": 491000
},
{
"epoch": 1.75,
"learning_rate": 0.0005,
"loss": 1.8824,
"step": 491500
},
{
"epoch": 1.75,
"learning_rate": 0.0005,
"loss": 1.8949,
"step": 492000
},
{
"epoch": 1.75,
"learning_rate": 0.0005,
"loss": 1.8723,
"step": 492500
},
{
"epoch": 1.75,
"learning_rate": 0.0005,
"loss": 1.8601,
"step": 493000
},
{
"epoch": 1.76,
"learning_rate": 0.0005,
"loss": 1.8848,
"step": 493500
},
{
"epoch": 1.76,
"learning_rate": 0.0005,
"loss": 1.8977,
"step": 494000
},
{
"epoch": 1.76,
"learning_rate": 0.0005,
"loss": 1.8697,
"step": 494500
},
{
"epoch": 1.76,
"learning_rate": 0.0005,
"loss": 1.8754,
"step": 495000
},
{
"epoch": 1.76,
"learning_rate": 0.0005,
"loss": 1.8617,
"step": 495500
},
{
"epoch": 1.76,
"learning_rate": 0.0005,
"loss": 1.8668,
"step": 496000
},
{
"epoch": 1.77,
"learning_rate": 0.0005,
"loss": 1.8823,
"step": 496500
},
{
"epoch": 1.77,
"learning_rate": 0.0005,
"loss": 1.8749,
"step": 497000
},
{
"epoch": 1.77,
"learning_rate": 0.0005,
"loss": 1.8653,
"step": 497500
},
{
"epoch": 1.77,
"learning_rate": 0.0005,
"loss": 1.8878,
"step": 498000
},
{
"epoch": 1.77,
"learning_rate": 0.0005,
"loss": 1.878,
"step": 498500
},
{
"epoch": 1.77,
"learning_rate": 0.0005,
"loss": 1.8723,
"step": 499000
},
{
"epoch": 1.78,
"learning_rate": 0.0005,
"loss": 1.8935,
"step": 499500
},
{
"epoch": 1.78,
"learning_rate": 0.0005,
"loss": 1.898,
"step": 500000
},
{
"epoch": 1.78,
"eval_bleu": 14.437,
"eval_gen_len": 17.2627,
"eval_runtime": 207.3898,
"eval_samples_per_second": 14.466,
"eval_steps_per_second": 1.808,
"step": 500000
},
{
"epoch": 1.78,
"learning_rate": 0.0005,
"loss": 1.8867,
"step": 500500
},
{
"epoch": 1.78,
"learning_rate": 0.0005,
"loss": 1.8817,
"step": 501000
},
{
"epoch": 1.78,
"learning_rate": 0.0005,
"loss": 1.8661,
"step": 501500
},
{
"epoch": 1.79,
"learning_rate": 0.0005,
"loss": 1.8818,
"step": 502000
},
{
"epoch": 1.79,
"learning_rate": 0.0005,
"loss": 1.8633,
"step": 502500
},
{
"epoch": 1.79,
"learning_rate": 0.0005,
"loss": 1.8829,
"step": 503000
},
{
"epoch": 1.79,
"learning_rate": 0.0005,
"loss": 1.8839,
"step": 503500
},
{
"epoch": 1.79,
"learning_rate": 0.0005,
"loss": 1.8755,
"step": 504000
},
{
"epoch": 1.79,
"learning_rate": 0.0005,
"loss": 1.8812,
"step": 504500
},
{
"epoch": 1.8,
"learning_rate": 0.0005,
"loss": 1.8726,
"step": 505000
},
{
"epoch": 1.8,
"learning_rate": 0.0005,
"loss": 1.8681,
"step": 505500
},
{
"epoch": 1.8,
"learning_rate": 0.0005,
"loss": 1.89,
"step": 506000
},
{
"epoch": 1.8,
"learning_rate": 0.0005,
"loss": 1.8702,
"step": 506500
},
{
"epoch": 1.8,
"learning_rate": 0.0005,
"loss": 1.8828,
"step": 507000
},
{
"epoch": 1.8,
"learning_rate": 0.0005,
"loss": 1.8632,
"step": 507500
},
{
"epoch": 1.81,
"learning_rate": 0.0005,
"loss": 1.89,
"step": 508000
},
{
"epoch": 1.81,
"learning_rate": 0.0005,
"loss": 1.8465,
"step": 508500
},
{
"epoch": 1.81,
"learning_rate": 0.0005,
"loss": 1.8747,
"step": 509000
},
{
"epoch": 1.81,
"learning_rate": 0.0005,
"loss": 1.8718,
"step": 509500
},
{
"epoch": 1.81,
"learning_rate": 0.0005,
"loss": 1.8686,
"step": 510000
},
{
"epoch": 1.81,
"eval_bleu": 14.1132,
"eval_gen_len": 17.263,
"eval_runtime": 207.7131,
"eval_samples_per_second": 14.443,
"eval_steps_per_second": 1.805,
"step": 510000
}
],
"max_steps": 1408995,
"num_train_epochs": 5,
"total_flos": 6.566535498761687e+17,
"trial_name": null,
"trial_params": null
}