| { | |
| "best_metric": 0.22902172803878784, | |
| "best_model_checkpoint": "./checkpoint-11500", | |
| "epoch": 15.0, | |
| "global_step": 11670, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.2933333333333334e-05, | |
| "loss": 14.1317, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.6266666666666667e-05, | |
| "loss": 5.2635, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 3.8268, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 5.293333333333334e-05, | |
| "loss": 3.2266, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 6.626666666666666e-05, | |
| "loss": 3.0952, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 3.0981762409210205, | |
| "eval_runtime": 326.629, | |
| "eval_samples_per_second": 25.414, | |
| "eval_steps_per_second": 0.796, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.960000000000001e-05, | |
| "loss": 3.0583, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 9.293333333333333e-05, | |
| "loss": 3.0349, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00010626666666666667, | |
| "loss": 2.9355, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00011960000000000001, | |
| "loss": 2.3372, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00013293333333333333, | |
| "loss": 1.7975, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.7887413501739502, | |
| "eval_runtime": 341.1915, | |
| "eval_samples_per_second": 24.329, | |
| "eval_steps_per_second": 0.762, | |
| "eval_wer": 0.5651080072872386, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00014626666666666665, | |
| "loss": 1.6159, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0001596, | |
| "loss": 1.5287, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00017293333333333335, | |
| "loss": 1.4876, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.00018626666666666668, | |
| "loss": 1.4606, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0001996, | |
| "loss": 1.4138, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 0.523814857006073, | |
| "eval_runtime": 341.7084, | |
| "eval_samples_per_second": 24.293, | |
| "eval_steps_per_second": 0.761, | |
| "eval_wer": 0.43889997397414765, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.00019809242871189774, | |
| "loss": 1.4258, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.000196125860373648, | |
| "loss": 1.3778, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.00019415929203539823, | |
| "loss": 1.3518, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 0.00019219272369714848, | |
| "loss": 1.3692, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 0.00019022615535889875, | |
| "loss": 1.344, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_loss": 0.4774917662143707, | |
| "eval_runtime": 337.4262, | |
| "eval_samples_per_second": 24.601, | |
| "eval_steps_per_second": 0.771, | |
| "eval_wer": 0.4318209421358549, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.000188259587020649, | |
| "loss": 1.3167, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 0.00018629301868239921, | |
| "loss": 1.3126, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.00018432645034414946, | |
| "loss": 1.3161, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 0.0001823598820058997, | |
| "loss": 1.2738, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 0.00018039331366764995, | |
| "loss": 1.2737, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_loss": 0.46475061774253845, | |
| "eval_runtime": 336.0925, | |
| "eval_samples_per_second": 24.699, | |
| "eval_steps_per_second": 0.774, | |
| "eval_wer": 0.4074954454758393, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.0001784267453294002, | |
| "loss": 1.262, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.00017646017699115044, | |
| "loss": 1.2496, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0001744936086529007, | |
| "loss": 1.2612, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.00017252704031465093, | |
| "loss": 1.2197, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 0.0001705604719764012, | |
| "loss": 1.2554, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "eval_loss": 0.4068518280982971, | |
| "eval_runtime": 331.555, | |
| "eval_samples_per_second": 25.037, | |
| "eval_steps_per_second": 0.784, | |
| "eval_wer": 0.36781469593129174, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.00016859390363815145, | |
| "loss": 1.2204, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 0.0001666273352999017, | |
| "loss": 1.2113, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 0.00016466076696165194, | |
| "loss": 1.2053, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 0.00016269419862340216, | |
| "loss": 1.2019, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 0.0001607276302851524, | |
| "loss": 1.1996, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_loss": 0.3914338946342468, | |
| "eval_runtime": 333.5366, | |
| "eval_samples_per_second": 24.888, | |
| "eval_steps_per_second": 0.78, | |
| "eval_wer": 0.36679101240565626, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.00015876106194690265, | |
| "loss": 1.1903, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.0001567944936086529, | |
| "loss": 1.1866, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 0.00015482792527040314, | |
| "loss": 1.1678, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 0.0001528613569321534, | |
| "loss": 1.1854, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 0.00015089478859390363, | |
| "loss": 1.1427, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_loss": 0.36938655376434326, | |
| "eval_runtime": 326.2453, | |
| "eval_samples_per_second": 25.444, | |
| "eval_steps_per_second": 0.797, | |
| "eval_wer": 0.3571614470373905, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.0001489282202556539, | |
| "loss": 1.1393, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 0.00014696165191740415, | |
| "loss": 1.1182, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.0001449950835791544, | |
| "loss": 1.127, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 0.00014302851524090464, | |
| "loss": 1.1431, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 0.0001410619469026549, | |
| "loss": 1.1372, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "eval_loss": 0.3567572236061096, | |
| "eval_runtime": 325.458, | |
| "eval_samples_per_second": 25.506, | |
| "eval_steps_per_second": 0.799, | |
| "eval_wer": 0.3500824151990978, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.0001390953785644051, | |
| "loss": 1.1226, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.00013712881022615535, | |
| "loss": 1.1019, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 0.0001351622418879056, | |
| "loss": 1.1031, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 0.00013321533923303834, | |
| "loss": 1.0882, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 0.00013124877089478858, | |
| "loss": 1.0831, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "eval_loss": 0.3331395387649536, | |
| "eval_runtime": 327.2346, | |
| "eval_samples_per_second": 25.367, | |
| "eval_steps_per_second": 0.795, | |
| "eval_wer": 0.3253058037650733, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.00012928220255653886, | |
| "loss": 1.1039, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.0001273156342182891, | |
| "loss": 1.0683, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 0.00012534906588003935, | |
| "loss": 1.0773, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.0001233824975417896, | |
| "loss": 1.1002, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 0.00012141592920353984, | |
| "loss": 1.1074, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "eval_loss": 0.333199679851532, | |
| "eval_runtime": 329.8287, | |
| "eval_samples_per_second": 25.168, | |
| "eval_steps_per_second": 0.788, | |
| "eval_wer": 0.3352129782250369, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.00011944936086529008, | |
| "loss": 1.0341, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.00011748279252704033, | |
| "loss": 1.0812, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 0.00011551622418879056, | |
| "loss": 1.0709, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 0.00011354965585054081, | |
| "loss": 1.0843, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "learning_rate": 0.00011158308751229105, | |
| "loss": 1.0536, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "eval_loss": 0.3130946755409241, | |
| "eval_runtime": 327.9895, | |
| "eval_samples_per_second": 25.309, | |
| "eval_steps_per_second": 0.793, | |
| "eval_wer": 0.3151557213498742, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 0.0001096165191740413, | |
| "loss": 1.0239, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 0.00010764995083579154, | |
| "loss": 1.0383, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 0.00010568338249754179, | |
| "loss": 1.0157, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.00010371681415929205, | |
| "loss": 1.0128, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 0.0001017502458210423, | |
| "loss": 1.0248, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "eval_loss": 0.30239033699035645, | |
| "eval_runtime": 328.0553, | |
| "eval_samples_per_second": 25.304, | |
| "eval_steps_per_second": 0.793, | |
| "eval_wer": 0.3022642491541598, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 9.978367748279254e-05, | |
| "loss": 0.9989, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 9.781710914454277e-05, | |
| "loss": 1.0151, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 9.585054080629302e-05, | |
| "loss": 0.9914, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 9.388397246804326e-05, | |
| "loss": 0.9893, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 9.193706981317602e-05, | |
| "loss": 1.0075, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.2947603166103363, | |
| "eval_runtime": 326.8764, | |
| "eval_samples_per_second": 25.395, | |
| "eval_steps_per_second": 0.795, | |
| "eval_wer": 0.3028368179057864, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "learning_rate": 8.997050147492626e-05, | |
| "loss": 0.9851, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 8.800393313667651e-05, | |
| "loss": 0.973, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 8.605703048180925e-05, | |
| "loss": 0.9623, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "learning_rate": 8.40904621435595e-05, | |
| "loss": 0.9598, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 8.212389380530974e-05, | |
| "loss": 0.979, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "eval_loss": 0.27962473034858704, | |
| "eval_runtime": 329.4084, | |
| "eval_samples_per_second": 25.2, | |
| "eval_steps_per_second": 0.789, | |
| "eval_wer": 0.2852953934241346, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 8.015732546705999e-05, | |
| "loss": 0.9582, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 7.819075712881023e-05, | |
| "loss": 0.9467, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 7.622418879056048e-05, | |
| "loss": 0.907, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "learning_rate": 7.425762045231072e-05, | |
| "loss": 0.944, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 7.229105211406097e-05, | |
| "loss": 0.9594, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "eval_loss": 0.2719425559043884, | |
| "eval_runtime": 329.4201, | |
| "eval_samples_per_second": 25.199, | |
| "eval_steps_per_second": 0.789, | |
| "eval_wer": 0.2789103843150863, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 7.032448377581121e-05, | |
| "loss": 0.9559, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 6.835791543756146e-05, | |
| "loss": 0.9368, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "learning_rate": 6.63913470993117e-05, | |
| "loss": 0.9362, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "learning_rate": 6.442477876106195e-05, | |
| "loss": 0.9134, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 6.24582104228122e-05, | |
| "loss": 0.9172, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "eval_loss": 0.2620205879211426, | |
| "eval_runtime": 325.6384, | |
| "eval_samples_per_second": 25.491, | |
| "eval_steps_per_second": 0.798, | |
| "eval_wer": 0.26950637633382496, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 6.049164208456244e-05, | |
| "loss": 0.9179, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 5.8525073746312686e-05, | |
| "loss": 0.9157, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 5.655850540806293e-05, | |
| "loss": 0.8974, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 11.44, | |
| "learning_rate": 5.459193706981318e-05, | |
| "loss": 0.8994, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "learning_rate": 5.262536873156343e-05, | |
| "loss": 0.9047, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "eval_loss": 0.25371646881103516, | |
| "eval_runtime": 328.7713, | |
| "eval_samples_per_second": 25.249, | |
| "eval_steps_per_second": 0.791, | |
| "eval_wer": 0.2596165524420925, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.7, | |
| "learning_rate": 5.065880039331367e-05, | |
| "loss": 0.8816, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 4.869223205506391e-05, | |
| "loss": 0.9007, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 11.95, | |
| "learning_rate": 4.672566371681416e-05, | |
| "loss": 0.9035, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 12.08, | |
| "learning_rate": 4.475909537856441e-05, | |
| "loss": 0.8705, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 4.279252704031465e-05, | |
| "loss": 0.8777, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "eval_loss": 0.24379895627498627, | |
| "eval_runtime": 329.8391, | |
| "eval_samples_per_second": 25.167, | |
| "eval_steps_per_second": 0.788, | |
| "eval_wer": 0.25250281946733755, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 4.0825958702064895e-05, | |
| "loss": 0.8734, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 12.47, | |
| "learning_rate": 3.887905604719764e-05, | |
| "loss": 0.8886, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 3.691248770894789e-05, | |
| "loss": 0.8626, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 12.72, | |
| "learning_rate": 3.4945919370698134e-05, | |
| "loss": 0.8669, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 3.297935103244838e-05, | |
| "loss": 0.8629, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "eval_loss": 0.2408979833126068, | |
| "eval_runtime": 327.8875, | |
| "eval_samples_per_second": 25.317, | |
| "eval_steps_per_second": 0.793, | |
| "eval_wer": 0.24934501604927561, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "learning_rate": 3.1012782694198625e-05, | |
| "loss": 0.8489, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 13.11, | |
| "learning_rate": 2.904621435594887e-05, | |
| "loss": 0.8356, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 2.7079646017699116e-05, | |
| "loss": 0.8596, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 13.37, | |
| "learning_rate": 2.5113077679449358e-05, | |
| "loss": 0.8401, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 2.3146509341199607e-05, | |
| "loss": 0.8575, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "eval_loss": 0.2366442084312439, | |
| "eval_runtime": 327.7324, | |
| "eval_samples_per_second": 25.329, | |
| "eval_steps_per_second": 0.793, | |
| "eval_wer": 0.24396633989763164, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 13.62, | |
| "learning_rate": 2.1179941002949856e-05, | |
| "loss": 0.8343, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 1.9213372664700098e-05, | |
| "loss": 0.8308, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "learning_rate": 1.7246804326450343e-05, | |
| "loss": 0.8431, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 14.01, | |
| "learning_rate": 1.5280235988200592e-05, | |
| "loss": 0.8468, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 1.3313667649950836e-05, | |
| "loss": 0.8361, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "eval_loss": 0.23166431486606598, | |
| "eval_runtime": 329.2059, | |
| "eval_samples_per_second": 25.215, | |
| "eval_steps_per_second": 0.79, | |
| "eval_wer": 0.23848356033660104, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.27, | |
| "learning_rate": 1.1347099311701081e-05, | |
| "loss": 0.824, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 9.380530973451327e-06, | |
| "loss": 0.8252, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 7.4336283185840714e-06, | |
| "loss": 0.8286, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 14.65, | |
| "learning_rate": 5.467059980334317e-06, | |
| "loss": 0.8077, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 3.500491642084563e-06, | |
| "loss": 0.8126, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "eval_loss": 0.22902172803878784, | |
| "eval_runtime": 327.6124, | |
| "eval_samples_per_second": 25.338, | |
| "eval_steps_per_second": 0.794, | |
| "eval_wer": 0.23818860067667216, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 1.5339233038348083e-06, | |
| "loss": 0.8118, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 11670, | |
| "total_flos": 4.014305196020058e+19, | |
| "train_loss": 1.3307904394651542, | |
| "train_runtime": 23640.5221, | |
| "train_samples_per_second": 15.785, | |
| "train_steps_per_second": 0.494 | |
| } | |
| ], | |
| "max_steps": 11670, | |
| "num_train_epochs": 15, | |
| "total_flos": 4.014305196020058e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |