yt-special-batch4-2lr5-small / trainer_state.json
TheRains's picture
End of training
39bdf7c
raw
history blame
25.9 kB
{
"best_metric": 51.26775176707088,
"best_model_checkpoint": "./output/small/yt-special-batch4-2lr5-small/checkpoint-4000",
"epoch": 0.4294425835265825,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 8.000000000000001e-07,
"loss": 2.4733,
"step": 25
},
{
"epoch": 0.0,
"learning_rate": 1.8000000000000001e-06,
"loss": 2.1237,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 2.8000000000000003e-06,
"loss": 1.7178,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 3.8000000000000005e-06,
"loss": 1.3884,
"step": 100
},
{
"epoch": 0.01,
"learning_rate": 4.800000000000001e-06,
"loss": 1.2175,
"step": 125
},
{
"epoch": 0.01,
"learning_rate": 5.8e-06,
"loss": 1.2592,
"step": 150
},
{
"epoch": 0.02,
"learning_rate": 6.800000000000001e-06,
"loss": 1.0327,
"step": 175
},
{
"epoch": 0.02,
"learning_rate": 7.800000000000002e-06,
"loss": 0.9806,
"step": 200
},
{
"epoch": 0.02,
"learning_rate": 8.8e-06,
"loss": 0.9788,
"step": 225
},
{
"epoch": 0.02,
"learning_rate": 9.800000000000001e-06,
"loss": 0.9626,
"step": 250
},
{
"epoch": 0.02,
"learning_rate": 1.0800000000000002e-05,
"loss": 1.0818,
"step": 275
},
{
"epoch": 0.03,
"learning_rate": 1.18e-05,
"loss": 0.909,
"step": 300
},
{
"epoch": 0.03,
"learning_rate": 1.2800000000000001e-05,
"loss": 0.9294,
"step": 325
},
{
"epoch": 0.03,
"learning_rate": 1.38e-05,
"loss": 0.8925,
"step": 350
},
{
"epoch": 0.03,
"learning_rate": 1.48e-05,
"loss": 0.9469,
"step": 375
},
{
"epoch": 0.03,
"learning_rate": 1.58e-05,
"loss": 0.9266,
"step": 400
},
{
"epoch": 0.04,
"learning_rate": 1.6800000000000002e-05,
"loss": 1.1455,
"step": 425
},
{
"epoch": 0.04,
"learning_rate": 1.7800000000000002e-05,
"loss": 1.0258,
"step": 450
},
{
"epoch": 0.04,
"learning_rate": 1.88e-05,
"loss": 1.1049,
"step": 475
},
{
"epoch": 0.04,
"learning_rate": 1.98e-05,
"loss": 0.9272,
"step": 500
},
{
"epoch": 0.05,
"learning_rate": 1.9911111111111112e-05,
"loss": 0.9699,
"step": 525
},
{
"epoch": 0.05,
"learning_rate": 1.98e-05,
"loss": 1.1994,
"step": 550
},
{
"epoch": 0.05,
"learning_rate": 1.968888888888889e-05,
"loss": 1.1111,
"step": 575
},
{
"epoch": 0.05,
"learning_rate": 1.957777777777778e-05,
"loss": 1.0309,
"step": 600
},
{
"epoch": 0.05,
"learning_rate": 1.9466666666666668e-05,
"loss": 0.8966,
"step": 625
},
{
"epoch": 0.06,
"learning_rate": 1.9355555555555556e-05,
"loss": 0.9403,
"step": 650
},
{
"epoch": 0.06,
"learning_rate": 1.9244444444444444e-05,
"loss": 1.1476,
"step": 675
},
{
"epoch": 0.06,
"learning_rate": 1.9133333333333335e-05,
"loss": 0.9617,
"step": 700
},
{
"epoch": 0.06,
"learning_rate": 1.9022222222222223e-05,
"loss": 1.159,
"step": 725
},
{
"epoch": 0.06,
"learning_rate": 1.8911111111111115e-05,
"loss": 1.2262,
"step": 750
},
{
"epoch": 0.07,
"learning_rate": 1.88e-05,
"loss": 1.0047,
"step": 775
},
{
"epoch": 0.07,
"learning_rate": 1.868888888888889e-05,
"loss": 1.0133,
"step": 800
},
{
"epoch": 0.07,
"learning_rate": 1.857777777777778e-05,
"loss": 0.8682,
"step": 825
},
{
"epoch": 0.07,
"learning_rate": 1.8466666666666667e-05,
"loss": 0.8822,
"step": 850
},
{
"epoch": 0.08,
"learning_rate": 1.835555555555556e-05,
"loss": 1.0939,
"step": 875
},
{
"epoch": 0.08,
"learning_rate": 1.8244444444444447e-05,
"loss": 1.0809,
"step": 900
},
{
"epoch": 0.08,
"learning_rate": 1.8133333333333335e-05,
"loss": 0.9912,
"step": 925
},
{
"epoch": 0.08,
"learning_rate": 1.8022222222222223e-05,
"loss": 1.0662,
"step": 950
},
{
"epoch": 0.08,
"learning_rate": 1.791111111111111e-05,
"loss": 0.9653,
"step": 975
},
{
"epoch": 0.09,
"learning_rate": 1.7800000000000002e-05,
"loss": 1.134,
"step": 1000
},
{
"epoch": 0.09,
"eval_loss": 1.0406965017318726,
"eval_runtime": 2372.0302,
"eval_samples_per_second": 1.033,
"eval_steps_per_second": 0.517,
"eval_wer": 97.97678490370274,
"step": 1000
},
{
"epoch": 0.09,
"learning_rate": 1.768888888888889e-05,
"loss": 0.9942,
"step": 1025
},
{
"epoch": 0.09,
"learning_rate": 1.7577777777777782e-05,
"loss": 0.9569,
"step": 1050
},
{
"epoch": 0.09,
"learning_rate": 1.7466666666666667e-05,
"loss": 1.1603,
"step": 1075
},
{
"epoch": 0.09,
"learning_rate": 1.7355555555555558e-05,
"loss": 0.9834,
"step": 1100
},
{
"epoch": 0.1,
"learning_rate": 1.7244444444444446e-05,
"loss": 0.9861,
"step": 1125
},
{
"epoch": 0.1,
"learning_rate": 1.7133333333333334e-05,
"loss": 0.9827,
"step": 1150
},
{
"epoch": 0.1,
"learning_rate": 1.7022222222222226e-05,
"loss": 1.0988,
"step": 1175
},
{
"epoch": 0.1,
"learning_rate": 1.691111111111111e-05,
"loss": 1.1107,
"step": 1200
},
{
"epoch": 0.11,
"learning_rate": 1.6800000000000002e-05,
"loss": 1.0515,
"step": 1225
},
{
"epoch": 0.11,
"learning_rate": 1.668888888888889e-05,
"loss": 0.9049,
"step": 1250
},
{
"epoch": 0.11,
"learning_rate": 1.6577777777777778e-05,
"loss": 1.0822,
"step": 1275
},
{
"epoch": 0.11,
"learning_rate": 1.646666666666667e-05,
"loss": 0.9417,
"step": 1300
},
{
"epoch": 0.11,
"learning_rate": 1.6355555555555557e-05,
"loss": 1.0671,
"step": 1325
},
{
"epoch": 0.12,
"learning_rate": 1.6244444444444446e-05,
"loss": 0.9822,
"step": 1350
},
{
"epoch": 0.12,
"learning_rate": 1.6133333333333334e-05,
"loss": 0.9884,
"step": 1375
},
{
"epoch": 0.12,
"learning_rate": 1.6022222222222225e-05,
"loss": 0.8488,
"step": 1400
},
{
"epoch": 0.12,
"learning_rate": 1.5911111111111113e-05,
"loss": 0.9872,
"step": 1425
},
{
"epoch": 0.12,
"learning_rate": 1.58e-05,
"loss": 0.9738,
"step": 1450
},
{
"epoch": 0.13,
"learning_rate": 1.5688888888888893e-05,
"loss": 1.026,
"step": 1475
},
{
"epoch": 0.13,
"learning_rate": 1.5577777777777777e-05,
"loss": 0.9435,
"step": 1500
},
{
"epoch": 0.13,
"learning_rate": 1.546666666666667e-05,
"loss": 0.9983,
"step": 1525
},
{
"epoch": 0.13,
"learning_rate": 1.5355555555555557e-05,
"loss": 0.8297,
"step": 1550
},
{
"epoch": 0.14,
"learning_rate": 1.5244444444444447e-05,
"loss": 0.9563,
"step": 1575
},
{
"epoch": 0.14,
"learning_rate": 1.5133333333333335e-05,
"loss": 0.9179,
"step": 1600
},
{
"epoch": 0.14,
"learning_rate": 1.5022222222222223e-05,
"loss": 0.8748,
"step": 1625
},
{
"epoch": 0.14,
"learning_rate": 1.4911111111111113e-05,
"loss": 0.8555,
"step": 1650
},
{
"epoch": 0.14,
"learning_rate": 1.48e-05,
"loss": 0.9714,
"step": 1675
},
{
"epoch": 0.15,
"learning_rate": 1.468888888888889e-05,
"loss": 0.8475,
"step": 1700
},
{
"epoch": 0.15,
"learning_rate": 1.457777777777778e-05,
"loss": 0.9448,
"step": 1725
},
{
"epoch": 0.15,
"learning_rate": 1.4466666666666668e-05,
"loss": 0.928,
"step": 1750
},
{
"epoch": 0.15,
"learning_rate": 1.4355555555555556e-05,
"loss": 1.0089,
"step": 1775
},
{
"epoch": 0.15,
"learning_rate": 1.4244444444444444e-05,
"loss": 0.945,
"step": 1800
},
{
"epoch": 0.16,
"learning_rate": 1.4133333333333334e-05,
"loss": 1.1229,
"step": 1825
},
{
"epoch": 0.16,
"learning_rate": 1.4022222222222224e-05,
"loss": 0.8683,
"step": 1850
},
{
"epoch": 0.16,
"learning_rate": 1.3911111111111114e-05,
"loss": 0.8754,
"step": 1875
},
{
"epoch": 0.16,
"learning_rate": 1.38e-05,
"loss": 0.8136,
"step": 1900
},
{
"epoch": 0.17,
"learning_rate": 1.368888888888889e-05,
"loss": 0.9256,
"step": 1925
},
{
"epoch": 0.17,
"learning_rate": 1.3577777777777778e-05,
"loss": 0.9101,
"step": 1950
},
{
"epoch": 0.17,
"learning_rate": 1.3466666666666668e-05,
"loss": 0.9439,
"step": 1975
},
{
"epoch": 0.17,
"learning_rate": 1.3355555555555557e-05,
"loss": 0.8923,
"step": 2000
},
{
"epoch": 0.17,
"eval_loss": 0.9184550642967224,
"eval_runtime": 2309.7182,
"eval_samples_per_second": 1.061,
"eval_steps_per_second": 0.531,
"eval_wer": 89.05388755593022,
"step": 2000
},
{
"epoch": 0.17,
"learning_rate": 1.3244444444444447e-05,
"loss": 0.9319,
"step": 2025
},
{
"epoch": 0.18,
"learning_rate": 1.3133333333333334e-05,
"loss": 1.0922,
"step": 2050
},
{
"epoch": 0.18,
"learning_rate": 1.3022222222222223e-05,
"loss": 0.9435,
"step": 2075
},
{
"epoch": 0.18,
"learning_rate": 1.2911111111111111e-05,
"loss": 0.846,
"step": 2100
},
{
"epoch": 0.18,
"learning_rate": 1.2800000000000001e-05,
"loss": 0.9983,
"step": 2125
},
{
"epoch": 0.18,
"learning_rate": 1.2688888888888891e-05,
"loss": 1.0326,
"step": 2150
},
{
"epoch": 0.19,
"learning_rate": 1.257777777777778e-05,
"loss": 0.8463,
"step": 2175
},
{
"epoch": 0.19,
"learning_rate": 1.2466666666666667e-05,
"loss": 0.8518,
"step": 2200
},
{
"epoch": 0.19,
"learning_rate": 1.2355555555555557e-05,
"loss": 0.8916,
"step": 2225
},
{
"epoch": 0.19,
"learning_rate": 1.2244444444444445e-05,
"loss": 0.8523,
"step": 2250
},
{
"epoch": 0.2,
"learning_rate": 1.2133333333333335e-05,
"loss": 1.0124,
"step": 2275
},
{
"epoch": 0.2,
"learning_rate": 1.2022222222222224e-05,
"loss": 1.1256,
"step": 2300
},
{
"epoch": 0.2,
"learning_rate": 1.191111111111111e-05,
"loss": 0.9919,
"step": 2325
},
{
"epoch": 0.2,
"learning_rate": 1.18e-05,
"loss": 1.1237,
"step": 2350
},
{
"epoch": 0.2,
"learning_rate": 1.168888888888889e-05,
"loss": 0.9261,
"step": 2375
},
{
"epoch": 0.21,
"learning_rate": 1.1577777777777778e-05,
"loss": 0.79,
"step": 2400
},
{
"epoch": 0.21,
"learning_rate": 1.1466666666666668e-05,
"loss": 0.9099,
"step": 2425
},
{
"epoch": 0.21,
"learning_rate": 1.1355555555555558e-05,
"loss": 0.8803,
"step": 2450
},
{
"epoch": 0.21,
"learning_rate": 1.1244444444444444e-05,
"loss": 0.9651,
"step": 2475
},
{
"epoch": 0.21,
"learning_rate": 1.1133333333333334e-05,
"loss": 0.9633,
"step": 2500
},
{
"epoch": 0.22,
"learning_rate": 1.1022222222222224e-05,
"loss": 0.7837,
"step": 2525
},
{
"epoch": 0.22,
"learning_rate": 1.0911111111111112e-05,
"loss": 0.7717,
"step": 2550
},
{
"epoch": 0.22,
"learning_rate": 1.0800000000000002e-05,
"loss": 0.7673,
"step": 2575
},
{
"epoch": 0.22,
"learning_rate": 1.0688888888888891e-05,
"loss": 0.9631,
"step": 2600
},
{
"epoch": 0.23,
"learning_rate": 1.0577777777777778e-05,
"loss": 0.9455,
"step": 2625
},
{
"epoch": 0.23,
"learning_rate": 1.0466666666666668e-05,
"loss": 0.7898,
"step": 2650
},
{
"epoch": 0.23,
"learning_rate": 1.0355555555555557e-05,
"loss": 1.0117,
"step": 2675
},
{
"epoch": 0.23,
"learning_rate": 1.0244444444444445e-05,
"loss": 0.771,
"step": 2700
},
{
"epoch": 0.23,
"learning_rate": 1.0133333333333335e-05,
"loss": 0.8904,
"step": 2725
},
{
"epoch": 0.24,
"learning_rate": 1.0022222222222222e-05,
"loss": 0.8569,
"step": 2750
},
{
"epoch": 0.24,
"learning_rate": 9.911111111111113e-06,
"loss": 0.8014,
"step": 2775
},
{
"epoch": 0.24,
"learning_rate": 9.800000000000001e-06,
"loss": 0.8306,
"step": 2800
},
{
"epoch": 0.24,
"learning_rate": 9.688888888888889e-06,
"loss": 0.7658,
"step": 2825
},
{
"epoch": 0.24,
"learning_rate": 9.577777777777779e-06,
"loss": 1.0627,
"step": 2850
},
{
"epoch": 0.25,
"learning_rate": 9.466666666666667e-06,
"loss": 0.9494,
"step": 2875
},
{
"epoch": 0.25,
"learning_rate": 9.355555555555557e-06,
"loss": 0.7214,
"step": 2900
},
{
"epoch": 0.25,
"learning_rate": 9.244444444444445e-06,
"loss": 0.9412,
"step": 2925
},
{
"epoch": 0.25,
"learning_rate": 9.133333333333335e-06,
"loss": 0.8554,
"step": 2950
},
{
"epoch": 0.26,
"learning_rate": 9.022222222222223e-06,
"loss": 0.7035,
"step": 2975
},
{
"epoch": 0.26,
"learning_rate": 8.91111111111111e-06,
"loss": 0.9713,
"step": 3000
},
{
"epoch": 0.26,
"eval_loss": 0.8536239266395569,
"eval_runtime": 2093.2662,
"eval_samples_per_second": 1.171,
"eval_steps_per_second": 0.586,
"eval_wer": 58.91317035211724,
"step": 3000
},
{
"epoch": 0.26,
"learning_rate": 8.8e-06,
"loss": 0.9826,
"step": 3025
},
{
"epoch": 0.26,
"learning_rate": 8.68888888888889e-06,
"loss": 1.0122,
"step": 3050
},
{
"epoch": 0.26,
"learning_rate": 8.577777777777778e-06,
"loss": 0.8062,
"step": 3075
},
{
"epoch": 0.27,
"learning_rate": 8.466666666666668e-06,
"loss": 0.8074,
"step": 3100
},
{
"epoch": 0.27,
"learning_rate": 8.355555555555556e-06,
"loss": 0.8672,
"step": 3125
},
{
"epoch": 0.27,
"learning_rate": 8.244444444444444e-06,
"loss": 0.9034,
"step": 3150
},
{
"epoch": 0.27,
"learning_rate": 8.133333333333334e-06,
"loss": 0.8459,
"step": 3175
},
{
"epoch": 0.27,
"learning_rate": 8.022222222222222e-06,
"loss": 0.8447,
"step": 3200
},
{
"epoch": 0.28,
"learning_rate": 7.911111111111112e-06,
"loss": 0.684,
"step": 3225
},
{
"epoch": 0.28,
"learning_rate": 7.800000000000002e-06,
"loss": 0.8424,
"step": 3250
},
{
"epoch": 0.28,
"learning_rate": 7.68888888888889e-06,
"loss": 0.8641,
"step": 3275
},
{
"epoch": 0.28,
"learning_rate": 7.5777777777777785e-06,
"loss": 0.8374,
"step": 3300
},
{
"epoch": 0.29,
"learning_rate": 7.4666666666666675e-06,
"loss": 0.7913,
"step": 3325
},
{
"epoch": 0.29,
"learning_rate": 7.3555555555555555e-06,
"loss": 0.863,
"step": 3350
},
{
"epoch": 0.29,
"learning_rate": 7.244444444444445e-06,
"loss": 0.7957,
"step": 3375
},
{
"epoch": 0.29,
"learning_rate": 7.133333333333334e-06,
"loss": 0.7969,
"step": 3400
},
{
"epoch": 0.29,
"learning_rate": 7.022222222222222e-06,
"loss": 0.825,
"step": 3425
},
{
"epoch": 0.3,
"learning_rate": 6.911111111111112e-06,
"loss": 0.7502,
"step": 3450
},
{
"epoch": 0.3,
"learning_rate": 6.800000000000001e-06,
"loss": 0.8589,
"step": 3475
},
{
"epoch": 0.3,
"learning_rate": 6.688888888888889e-06,
"loss": 0.6825,
"step": 3500
},
{
"epoch": 0.3,
"learning_rate": 6.577777777777779e-06,
"loss": 0.9456,
"step": 3525
},
{
"epoch": 0.3,
"learning_rate": 6.466666666666667e-06,
"loss": 0.8527,
"step": 3550
},
{
"epoch": 0.31,
"learning_rate": 6.355555555555556e-06,
"loss": 0.8028,
"step": 3575
},
{
"epoch": 0.31,
"learning_rate": 6.2444444444444456e-06,
"loss": 0.6831,
"step": 3600
},
{
"epoch": 0.31,
"learning_rate": 6.133333333333334e-06,
"loss": 0.787,
"step": 3625
},
{
"epoch": 0.31,
"learning_rate": 6.0222222222222225e-06,
"loss": 0.8547,
"step": 3650
},
{
"epoch": 0.32,
"learning_rate": 5.911111111111112e-06,
"loss": 0.8395,
"step": 3675
},
{
"epoch": 0.32,
"learning_rate": 5.8e-06,
"loss": 0.9031,
"step": 3700
},
{
"epoch": 0.32,
"learning_rate": 5.688888888888889e-06,
"loss": 0.7951,
"step": 3725
},
{
"epoch": 0.32,
"learning_rate": 5.577777777777778e-06,
"loss": 0.7177,
"step": 3750
},
{
"epoch": 0.32,
"learning_rate": 5.466666666666667e-06,
"loss": 0.7743,
"step": 3775
},
{
"epoch": 0.33,
"learning_rate": 5.355555555555556e-06,
"loss": 0.7501,
"step": 3800
},
{
"epoch": 0.33,
"learning_rate": 5.244444444444445e-06,
"loss": 0.8543,
"step": 3825
},
{
"epoch": 0.33,
"learning_rate": 5.133333333333334e-06,
"loss": 0.8073,
"step": 3850
},
{
"epoch": 0.33,
"learning_rate": 5.022222222222223e-06,
"loss": 0.8463,
"step": 3875
},
{
"epoch": 0.33,
"learning_rate": 4.911111111111112e-06,
"loss": 0.8744,
"step": 3900
},
{
"epoch": 0.34,
"learning_rate": 4.800000000000001e-06,
"loss": 0.8423,
"step": 3925
},
{
"epoch": 0.34,
"learning_rate": 4.6888888888888895e-06,
"loss": 0.616,
"step": 3950
},
{
"epoch": 0.34,
"learning_rate": 4.5777777777777785e-06,
"loss": 0.8428,
"step": 3975
},
{
"epoch": 0.34,
"learning_rate": 4.4666666666666665e-06,
"loss": 0.7834,
"step": 4000
},
{
"epoch": 0.34,
"eval_loss": 0.7838221192359924,
"eval_runtime": 2041.0552,
"eval_samples_per_second": 1.201,
"eval_steps_per_second": 0.601,
"eval_wer": 51.26775176707088,
"step": 4000
},
{
"epoch": 0.35,
"learning_rate": 4.3555555555555555e-06,
"loss": 0.7795,
"step": 4025
},
{
"epoch": 0.35,
"learning_rate": 4.244444444444445e-06,
"loss": 0.7758,
"step": 4050
},
{
"epoch": 0.35,
"learning_rate": 4.133333333333333e-06,
"loss": 0.7385,
"step": 4075
},
{
"epoch": 0.35,
"learning_rate": 4.022222222222222e-06,
"loss": 0.9365,
"step": 4100
},
{
"epoch": 0.35,
"learning_rate": 3.911111111111112e-06,
"loss": 0.7566,
"step": 4125
},
{
"epoch": 0.36,
"learning_rate": 3.8000000000000005e-06,
"loss": 0.8556,
"step": 4150
},
{
"epoch": 0.36,
"learning_rate": 3.688888888888889e-06,
"loss": 0.6934,
"step": 4175
},
{
"epoch": 0.36,
"learning_rate": 3.577777777777778e-06,
"loss": 0.7919,
"step": 4200
},
{
"epoch": 0.36,
"learning_rate": 3.4666666666666672e-06,
"loss": 0.7696,
"step": 4225
},
{
"epoch": 0.37,
"learning_rate": 3.3555555555555557e-06,
"loss": 0.7166,
"step": 4250
},
{
"epoch": 0.37,
"learning_rate": 3.2444444444444446e-06,
"loss": 0.67,
"step": 4275
},
{
"epoch": 0.37,
"learning_rate": 3.137777777777778e-06,
"loss": 0.9282,
"step": 4300
},
{
"epoch": 0.37,
"learning_rate": 3.0266666666666668e-06,
"loss": 0.7539,
"step": 4325
},
{
"epoch": 0.37,
"learning_rate": 2.915555555555556e-06,
"loss": 0.7936,
"step": 4350
},
{
"epoch": 0.38,
"learning_rate": 2.8044444444444446e-06,
"loss": 0.8151,
"step": 4375
},
{
"epoch": 0.38,
"learning_rate": 2.6933333333333335e-06,
"loss": 0.7438,
"step": 4400
},
{
"epoch": 0.38,
"learning_rate": 2.5822222222222224e-06,
"loss": 0.8626,
"step": 4425
},
{
"epoch": 0.38,
"learning_rate": 2.4711111111111114e-06,
"loss": 0.6498,
"step": 4450
},
{
"epoch": 0.38,
"learning_rate": 2.3600000000000003e-06,
"loss": 0.7034,
"step": 4475
},
{
"epoch": 0.39,
"learning_rate": 2.248888888888889e-06,
"loss": 0.7376,
"step": 4500
},
{
"epoch": 0.39,
"learning_rate": 2.137777777777778e-06,
"loss": 0.7738,
"step": 4525
},
{
"epoch": 0.39,
"learning_rate": 2.0266666666666666e-06,
"loss": 0.5934,
"step": 4550
},
{
"epoch": 0.39,
"learning_rate": 1.915555555555556e-06,
"loss": 0.8443,
"step": 4575
},
{
"epoch": 0.4,
"learning_rate": 1.8044444444444444e-06,
"loss": 0.77,
"step": 4600
},
{
"epoch": 0.4,
"learning_rate": 1.6933333333333336e-06,
"loss": 0.7721,
"step": 4625
},
{
"epoch": 0.4,
"learning_rate": 1.5822222222222223e-06,
"loss": 0.8138,
"step": 4650
},
{
"epoch": 0.4,
"learning_rate": 1.4711111111111112e-06,
"loss": 0.748,
"step": 4675
},
{
"epoch": 0.4,
"learning_rate": 1.3600000000000001e-06,
"loss": 0.9393,
"step": 4700
},
{
"epoch": 0.41,
"learning_rate": 1.248888888888889e-06,
"loss": 0.8364,
"step": 4725
},
{
"epoch": 0.41,
"learning_rate": 1.137777777777778e-06,
"loss": 0.7931,
"step": 4750
},
{
"epoch": 0.41,
"learning_rate": 1.0266666666666669e-06,
"loss": 0.6851,
"step": 4775
},
{
"epoch": 0.41,
"learning_rate": 9.155555555555557e-07,
"loss": 0.7286,
"step": 4800
},
{
"epoch": 0.41,
"learning_rate": 8.044444444444445e-07,
"loss": 0.741,
"step": 4825
},
{
"epoch": 0.42,
"learning_rate": 6.933333333333334e-07,
"loss": 0.7786,
"step": 4850
},
{
"epoch": 0.42,
"learning_rate": 5.822222222222223e-07,
"loss": 0.8268,
"step": 4875
},
{
"epoch": 0.42,
"learning_rate": 4.7111111111111113e-07,
"loss": 0.6262,
"step": 4900
},
{
"epoch": 0.42,
"learning_rate": 3.6e-07,
"loss": 0.7651,
"step": 4925
},
{
"epoch": 0.43,
"learning_rate": 2.488888888888889e-07,
"loss": 0.659,
"step": 4950
},
{
"epoch": 0.43,
"learning_rate": 1.3777777777777778e-07,
"loss": 0.728,
"step": 4975
},
{
"epoch": 0.43,
"learning_rate": 2.6666666666666667e-08,
"loss": 0.78,
"step": 5000
},
{
"epoch": 0.43,
"eval_loss": 0.7437878847122192,
"eval_runtime": 2038.0057,
"eval_samples_per_second": 1.203,
"eval_steps_per_second": 0.602,
"eval_wer": 52.19505868620712,
"step": 5000
},
{
"epoch": 0.43,
"step": 5000,
"total_flos": 5.7717080064e+18,
"train_loss": 0.9145148426055908,
"train_runtime": 25154.7698,
"train_samples_per_second": 0.795,
"train_steps_per_second": 0.199
}
],
"max_steps": 5000,
"num_train_epochs": 1,
"total_flos": 5.7717080064e+18,
"trial_name": null,
"trial_params": null
}