| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.998632010943915, | |
| "global_step": 18250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3.7125e-06, | |
| "loss": 12.2134, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 7.4625e-06, | |
| "loss": 6.115, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.1212499999999998e-05, | |
| "loss": 4.1403, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.49625e-05, | |
| "loss": 3.6727, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.8712499999999997e-05, | |
| "loss": 3.3224, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 3.267564058303833, | |
| "eval_runtime": 144.3422, | |
| "eval_samples_per_second": 33.552, | |
| "eval_steps_per_second": 4.198, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.2462499999999997e-05, | |
| "loss": 3.1389, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 2.6212499999999997e-05, | |
| "loss": 3.069, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.99625e-05, | |
| "loss": 3.0412, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 3.37125e-05, | |
| "loss": 2.9725, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 3.7462499999999996e-05, | |
| "loss": 2.9319, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_loss": 2.928657054901123, | |
| "eval_runtime": 144.855, | |
| "eval_samples_per_second": 33.433, | |
| "eval_steps_per_second": 4.183, | |
| "eval_wer": 1.0000283848992335, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.12125e-05, | |
| "loss": 2.9171, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 4.4962499999999995e-05, | |
| "loss": 2.8588, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 4.871249999999999e-05, | |
| "loss": 2.7606, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 5.2462499999999994e-05, | |
| "loss": 2.4333, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 5.62125e-05, | |
| "loss": 2.1173, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "eval_loss": 1.147773265838623, | |
| "eval_runtime": 146.3922, | |
| "eval_samples_per_second": 33.082, | |
| "eval_steps_per_second": 4.14, | |
| "eval_wer": 0.8787680953732614, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 5.9962499999999994e-05, | |
| "loss": 1.9512, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 6.37125e-05, | |
| "loss": 1.8501, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 6.746249999999999e-05, | |
| "loss": 1.7846, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 7.121249999999999e-05, | |
| "loss": 1.7365, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 7.49625e-05, | |
| "loss": 1.6973, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "eval_loss": 0.6748705506324768, | |
| "eval_runtime": 146.3732, | |
| "eval_samples_per_second": 33.087, | |
| "eval_steps_per_second": 4.14, | |
| "eval_wer": 0.6546693159239285, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 7.45476923076923e-05, | |
| "loss": 1.6824, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 7.408615384615384e-05, | |
| "loss": 1.6642, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 7.362461538461538e-05, | |
| "loss": 1.6098, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 7.316769230769229e-05, | |
| "loss": 1.6063, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 7.270615384615384e-05, | |
| "loss": 1.5865, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_loss": 0.5499770641326904, | |
| "eval_runtime": 146.214, | |
| "eval_samples_per_second": 33.123, | |
| "eval_steps_per_second": 4.145, | |
| "eval_wer": 0.5633550950894124, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 7.224461538461538e-05, | |
| "loss": 1.5758, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 7.178307692307692e-05, | |
| "loss": 1.5329, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 7.132153846153845e-05, | |
| "loss": 1.5395, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 7.086e-05, | |
| "loss": 1.5163, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 7.039846153846153e-05, | |
| "loss": 1.5094, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "eval_loss": 0.4839838743209839, | |
| "eval_runtime": 146.4688, | |
| "eval_samples_per_second": 33.065, | |
| "eval_steps_per_second": 4.137, | |
| "eval_wer": 0.542974737439682, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 6.993692307692307e-05, | |
| "loss": 1.4862, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 6.94753846153846e-05, | |
| "loss": 1.4738, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 6.901384615384615e-05, | |
| "loss": 1.4667, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 6.855230769230769e-05, | |
| "loss": 1.4564, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 6.809076923076922e-05, | |
| "loss": 1.4644, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "eval_loss": 0.48440465331077576, | |
| "eval_runtime": 146.4275, | |
| "eval_samples_per_second": 33.074, | |
| "eval_steps_per_second": 4.139, | |
| "eval_wer": 0.4142492194152711, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 6.762923076923076e-05, | |
| "loss": 1.4389, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 6.716769230769231e-05, | |
| "loss": 1.437, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 6.670615384615384e-05, | |
| "loss": 1.4278, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 6.624461538461537e-05, | |
| "loss": 1.4063, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 6.578307692307691e-05, | |
| "loss": 1.4061, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_loss": 0.4355711340904236, | |
| "eval_runtime": 146.0305, | |
| "eval_samples_per_second": 33.164, | |
| "eval_steps_per_second": 4.15, | |
| "eval_wer": 0.3808401930173148, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "learning_rate": 6.532153846153845e-05, | |
| "loss": 1.3829, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 6.486e-05, | |
| "loss": 1.3993, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 6.439846153846153e-05, | |
| "loss": 1.3875, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 6.393692307692306e-05, | |
| "loss": 1.4017, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 12.33, | |
| "learning_rate": 6.347538461538461e-05, | |
| "loss": 1.3584, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.33, | |
| "eval_loss": 0.41921573877334595, | |
| "eval_runtime": 145.7061, | |
| "eval_samples_per_second": 33.238, | |
| "eval_steps_per_second": 4.159, | |
| "eval_wer": 0.3697700823162078, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 6.301846153846153e-05, | |
| "loss": 1.3732, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 12.88, | |
| "learning_rate": 6.255692307692307e-05, | |
| "loss": 1.3719, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "learning_rate": 6.20953846153846e-05, | |
| "loss": 1.3742, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "learning_rate": 6.163384615384614e-05, | |
| "loss": 1.3416, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "learning_rate": 6.117230769230768e-05, | |
| "loss": 1.3438, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "eval_loss": 0.39804813265800476, | |
| "eval_runtime": 146.4763, | |
| "eval_samples_per_second": 33.063, | |
| "eval_steps_per_second": 4.137, | |
| "eval_wer": 0.3583877377235311, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.97, | |
| "learning_rate": 6.0710769230769224e-05, | |
| "loss": 1.3397, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 6.024923076923076e-05, | |
| "loss": 1.3228, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 5.97876923076923e-05, | |
| "loss": 1.3181, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 5.9326153846153835e-05, | |
| "loss": 1.3505, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "learning_rate": 5.886461538461538e-05, | |
| "loss": 1.3332, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "eval_loss": 0.3896285891532898, | |
| "eval_runtime": 145.3992, | |
| "eval_samples_per_second": 33.308, | |
| "eval_steps_per_second": 4.168, | |
| "eval_wer": 0.3571671870564859, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.34, | |
| "learning_rate": 5.840307692307692e-05, | |
| "loss": 1.3081, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 5.794153846153845e-05, | |
| "loss": 1.3305, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 15.89, | |
| "learning_rate": 5.747999999999999e-05, | |
| "loss": 1.3206, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 16.16, | |
| "learning_rate": 5.702307692307692e-05, | |
| "loss": 1.3205, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "learning_rate": 5.6561538461538456e-05, | |
| "loss": 1.3025, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "eval_loss": 0.3834877014160156, | |
| "eval_runtime": 145.719, | |
| "eval_samples_per_second": 33.235, | |
| "eval_steps_per_second": 4.159, | |
| "eval_wer": 0.34870848708487084, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.71, | |
| "learning_rate": 5.6099999999999995e-05, | |
| "loss": 1.3008, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 5.5638461538461534e-05, | |
| "loss": 1.2926, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 5.5176923076923067e-05, | |
| "loss": 1.2871, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 5.471538461538461e-05, | |
| "loss": 1.2688, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 5.425384615384615e-05, | |
| "loss": 1.2979, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "eval_loss": 0.37809377908706665, | |
| "eval_runtime": 145.3953, | |
| "eval_samples_per_second": 33.309, | |
| "eval_steps_per_second": 4.168, | |
| "eval_wer": 0.3416974169741697, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "learning_rate": 5.3792307692307684e-05, | |
| "loss": 1.2865, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 18.36, | |
| "learning_rate": 5.3330769230769223e-05, | |
| "loss": 1.2795, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 18.63, | |
| "learning_rate": 5.286923076923076e-05, | |
| "loss": 1.271, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 5.240769230769231e-05, | |
| "loss": 1.2595, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 5.194615384615384e-05, | |
| "loss": 1.2736, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "eval_loss": 0.37344905734062195, | |
| "eval_runtime": 145.9621, | |
| "eval_samples_per_second": 33.18, | |
| "eval_steps_per_second": 4.152, | |
| "eval_wer": 0.32699403917116093, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "learning_rate": 5.148461538461538e-05, | |
| "loss": 1.2619, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 5.102307692307692e-05, | |
| "loss": 1.2351, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 5.056153846153845e-05, | |
| "loss": 1.277, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 5.01e-05, | |
| "loss": 1.2559, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 20.55, | |
| "learning_rate": 4.963846153846154e-05, | |
| "loss": 1.2415, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.55, | |
| "eval_loss": 0.3636943995952606, | |
| "eval_runtime": 146.1853, | |
| "eval_samples_per_second": 33.129, | |
| "eval_steps_per_second": 4.145, | |
| "eval_wer": 0.3316491626454726, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.82, | |
| "learning_rate": 4.917692307692307e-05, | |
| "loss": 1.2383, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 21.1, | |
| "learning_rate": 4.871538461538461e-05, | |
| "loss": 1.2373, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 21.37, | |
| "learning_rate": 4.825384615384615e-05, | |
| "loss": 1.2197, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 21.64, | |
| "learning_rate": 4.7792307692307694e-05, | |
| "loss": 1.2291, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "learning_rate": 4.733076923076923e-05, | |
| "loss": 1.2255, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "eval_loss": 0.35461699962615967, | |
| "eval_runtime": 145.4805, | |
| "eval_samples_per_second": 33.29, | |
| "eval_steps_per_second": 4.166, | |
| "eval_wer": 0.3146749929037752, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 22.19, | |
| "learning_rate": 4.6869230769230766e-05, | |
| "loss": 1.2578, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 22.47, | |
| "learning_rate": 4.64076923076923e-05, | |
| "loss": 1.2145, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 22.74, | |
| "learning_rate": 4.594615384615384e-05, | |
| "loss": 1.2263, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 4.5484615384615384e-05, | |
| "loss": 1.2557, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 23.29, | |
| "learning_rate": 4.502307692307692e-05, | |
| "loss": 1.2193, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 23.29, | |
| "eval_loss": 0.35236233472824097, | |
| "eval_runtime": 145.9059, | |
| "eval_samples_per_second": 33.193, | |
| "eval_steps_per_second": 4.153, | |
| "eval_wer": 0.31964235026965654, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 23.56, | |
| "learning_rate": 4.456615384615384e-05, | |
| "loss": 1.2092, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 23.83, | |
| "learning_rate": 4.410461538461538e-05, | |
| "loss": 1.222, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 24.11, | |
| "learning_rate": 4.364307692307692e-05, | |
| "loss": 1.2004, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 24.38, | |
| "learning_rate": 4.318153846153846e-05, | |
| "loss": 1.2121, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 24.66, | |
| "learning_rate": 4.272e-05, | |
| "loss": 1.2104, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.66, | |
| "eval_loss": 0.3402980864048004, | |
| "eval_runtime": 146.1669, | |
| "eval_samples_per_second": 33.133, | |
| "eval_steps_per_second": 4.146, | |
| "eval_wer": 0.30967925063866025, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.93, | |
| "learning_rate": 4.225846153846153e-05, | |
| "loss": 1.2001, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 25.21, | |
| "learning_rate": 4.179692307692307e-05, | |
| "loss": 1.2004, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 25.48, | |
| "learning_rate": 4.1335384615384615e-05, | |
| "loss": 1.1803, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 25.75, | |
| "learning_rate": 4.0873846153846154e-05, | |
| "loss": 1.1934, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 26.03, | |
| "learning_rate": 4.041230769230769e-05, | |
| "loss": 1.1965, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 26.03, | |
| "eval_loss": 0.35078364610671997, | |
| "eval_runtime": 145.889, | |
| "eval_samples_per_second": 33.196, | |
| "eval_steps_per_second": 4.154, | |
| "eval_wer": 0.30931024694862336, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 26.3, | |
| "learning_rate": 3.9950769230769226e-05, | |
| "loss": 1.1909, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 26.57, | |
| "learning_rate": 3.948923076923076e-05, | |
| "loss": 1.1886, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 26.85, | |
| "learning_rate": 3.9027692307692305e-05, | |
| "loss": 1.1785, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "learning_rate": 3.8566153846153844e-05, | |
| "loss": 1.1675, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "learning_rate": 3.810461538461538e-05, | |
| "loss": 1.1976, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "eval_loss": 0.34191542863845825, | |
| "eval_runtime": 145.8469, | |
| "eval_samples_per_second": 33.206, | |
| "eval_steps_per_second": 4.155, | |
| "eval_wer": 0.30712460970763555, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 27.67, | |
| "learning_rate": 3.7643076923076916e-05, | |
| "loss": 1.1729, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 27.94, | |
| "learning_rate": 3.718153846153846e-05, | |
| "loss": 1.1852, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 28.22, | |
| "learning_rate": 3.6719999999999994e-05, | |
| "loss": 1.1881, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 28.49, | |
| "learning_rate": 3.625846153846153e-05, | |
| "loss": 1.1632, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "learning_rate": 3.580153846153846e-05, | |
| "loss": 1.182, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "eval_loss": 0.3363555371761322, | |
| "eval_runtime": 146.1933, | |
| "eval_samples_per_second": 33.127, | |
| "eval_steps_per_second": 4.145, | |
| "eval_wer": 0.2963383479988646, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 29.04, | |
| "learning_rate": 3.534e-05, | |
| "loss": 1.1535, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 29.31, | |
| "learning_rate": 3.4878461538461536e-05, | |
| "loss": 1.1636, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 29.59, | |
| "learning_rate": 3.4416923076923076e-05, | |
| "loss": 1.1496, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 29.86, | |
| "learning_rate": 3.3955384615384615e-05, | |
| "loss": 1.1507, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 30.14, | |
| "learning_rate": 3.349384615384615e-05, | |
| "loss": 1.158, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 30.14, | |
| "eval_loss": 0.333802729845047, | |
| "eval_runtime": 145.3999, | |
| "eval_samples_per_second": 33.308, | |
| "eval_steps_per_second": 4.168, | |
| "eval_wer": 0.2931876241839341, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 30.41, | |
| "learning_rate": 3.303230769230769e-05, | |
| "loss": 1.1527, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 30.68, | |
| "learning_rate": 3.2570769230769226e-05, | |
| "loss": 1.1484, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 30.96, | |
| "learning_rate": 3.2109230769230765e-05, | |
| "loss": 1.145, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 31.23, | |
| "learning_rate": 3.1647692307692304e-05, | |
| "loss": 1.1388, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 31.51, | |
| "learning_rate": 3.118615384615384e-05, | |
| "loss": 1.1414, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 31.51, | |
| "eval_loss": 0.3376125693321228, | |
| "eval_runtime": 147.5034, | |
| "eval_samples_per_second": 32.833, | |
| "eval_steps_per_second": 4.108, | |
| "eval_wer": 0.29398240136247517, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 31.78, | |
| "learning_rate": 3.072461538461538e-05, | |
| "loss": 1.1351, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 32.05, | |
| "learning_rate": 3.026307692307692e-05, | |
| "loss": 1.1228, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 32.33, | |
| "learning_rate": 2.980153846153846e-05, | |
| "loss": 1.1443, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 32.6, | |
| "learning_rate": 2.9339999999999997e-05, | |
| "loss": 1.1335, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 32.88, | |
| "learning_rate": 2.8878461538461536e-05, | |
| "loss": 1.1402, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 32.88, | |
| "eval_loss": 0.3370179235935211, | |
| "eval_runtime": 146.1673, | |
| "eval_samples_per_second": 33.133, | |
| "eval_steps_per_second": 4.146, | |
| "eval_wer": 0.2891285835935282, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 33.15, | |
| "learning_rate": 2.8416923076923075e-05, | |
| "loss": 1.1435, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 33.42, | |
| "learning_rate": 2.795538461538461e-05, | |
| "loss": 1.1122, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 33.7, | |
| "learning_rate": 2.7493846153846154e-05, | |
| "loss": 1.1107, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 33.97, | |
| "learning_rate": 2.703230769230769e-05, | |
| "loss": 1.1126, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 34.25, | |
| "learning_rate": 2.657076923076923e-05, | |
| "loss": 1.1213, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 34.25, | |
| "eval_loss": 0.3200843334197998, | |
| "eval_runtime": 146.0978, | |
| "eval_samples_per_second": 33.149, | |
| "eval_steps_per_second": 4.148, | |
| "eval_wer": 0.28742548963951176, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 34.52, | |
| "learning_rate": 2.6109230769230768e-05, | |
| "loss": 1.1115, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 34.79, | |
| "learning_rate": 2.5647692307692304e-05, | |
| "loss": 1.1069, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 35.07, | |
| "learning_rate": 2.5186153846153843e-05, | |
| "loss": 1.1154, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 35.34, | |
| "learning_rate": 2.4729230769230764e-05, | |
| "loss": 1.1079, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 35.62, | |
| "learning_rate": 2.4267692307692307e-05, | |
| "loss": 1.1207, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 35.62, | |
| "eval_loss": 0.3260834515094757, | |
| "eval_runtime": 147.1601, | |
| "eval_samples_per_second": 32.91, | |
| "eval_steps_per_second": 4.118, | |
| "eval_wer": 0.28262844166903206, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 35.89, | |
| "learning_rate": 2.3806153846153843e-05, | |
| "loss": 1.1034, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 36.16, | |
| "learning_rate": 2.3344615384615386e-05, | |
| "loss": 1.1157, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 36.44, | |
| "learning_rate": 2.288307692307692e-05, | |
| "loss": 1.0868, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 36.71, | |
| "learning_rate": 2.242153846153846e-05, | |
| "loss": 1.0985, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 36.98, | |
| "learning_rate": 2.196e-05, | |
| "loss": 1.1074, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 36.98, | |
| "eval_loss": 0.3117202818393707, | |
| "eval_runtime": 145.4969, | |
| "eval_samples_per_second": 33.286, | |
| "eval_steps_per_second": 4.165, | |
| "eval_wer": 0.2785977859778598, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 37.26, | |
| "learning_rate": 2.1498461538461536e-05, | |
| "loss": 1.0955, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 37.53, | |
| "learning_rate": 2.1036923076923075e-05, | |
| "loss": 1.0826, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 37.81, | |
| "learning_rate": 2.0575384615384614e-05, | |
| "loss": 1.0897, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 38.08, | |
| "learning_rate": 2.0113846153846153e-05, | |
| "loss": 1.0992, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 38.36, | |
| "learning_rate": 1.965230769230769e-05, | |
| "loss": 1.0818, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 38.36, | |
| "eval_loss": 0.3194292485713959, | |
| "eval_runtime": 145.4744, | |
| "eval_samples_per_second": 33.291, | |
| "eval_steps_per_second": 4.166, | |
| "eval_wer": 0.2775759296054499, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 38.63, | |
| "learning_rate": 1.919076923076923e-05, | |
| "loss": 1.0812, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 38.9, | |
| "learning_rate": 1.8729230769230768e-05, | |
| "loss": 1.0851, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 39.18, | |
| "learning_rate": 1.8267692307692307e-05, | |
| "loss": 1.0824, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 39.45, | |
| "learning_rate": 1.7806153846153843e-05, | |
| "loss": 1.1003, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 39.73, | |
| "learning_rate": 1.7344615384615382e-05, | |
| "loss": 1.0889, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 39.73, | |
| "eval_loss": 0.31878727674484253, | |
| "eval_runtime": 145.398, | |
| "eval_samples_per_second": 33.309, | |
| "eval_steps_per_second": 4.168, | |
| "eval_wer": 0.2738291229066137, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 1.688307692307692e-05, | |
| "loss": 1.0847, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 40.27, | |
| "learning_rate": 1.642153846153846e-05, | |
| "loss": 1.0707, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 40.55, | |
| "learning_rate": 1.5959999999999996e-05, | |
| "loss": 1.0664, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 40.82, | |
| "learning_rate": 1.5498461538461536e-05, | |
| "loss": 1.0814, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 41.1, | |
| "learning_rate": 1.5036923076923075e-05, | |
| "loss": 1.0672, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 41.1, | |
| "eval_loss": 0.31958457827568054, | |
| "eval_runtime": 145.6438, | |
| "eval_samples_per_second": 33.252, | |
| "eval_steps_per_second": 4.161, | |
| "eval_wer": 0.277348850411581, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 41.37, | |
| "learning_rate": 1.4575384615384614e-05, | |
| "loss": 1.0709, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 41.64, | |
| "learning_rate": 1.4113846153846153e-05, | |
| "loss": 1.0833, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 41.92, | |
| "learning_rate": 1.365230769230769e-05, | |
| "loss": 1.0694, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 42.19, | |
| "learning_rate": 1.3190769230769228e-05, | |
| "loss": 1.063, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 42.47, | |
| "learning_rate": 1.2729230769230768e-05, | |
| "loss": 1.0838, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 42.47, | |
| "eval_loss": 0.3129686713218689, | |
| "eval_runtime": 146.1643, | |
| "eval_samples_per_second": 33.134, | |
| "eval_steps_per_second": 4.146, | |
| "eval_wer": 0.2739142776043145, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 42.74, | |
| "learning_rate": 1.2267692307692307e-05, | |
| "loss": 1.0469, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 1.181076923076923e-05, | |
| "loss": 1.0746, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 43.29, | |
| "learning_rate": 1.1349230769230769e-05, | |
| "loss": 1.0495, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 43.56, | |
| "learning_rate": 1.0887692307692306e-05, | |
| "loss": 1.0553, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 43.83, | |
| "learning_rate": 1.0426153846153846e-05, | |
| "loss": 1.0553, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 43.83, | |
| "eval_loss": 0.3164559602737427, | |
| "eval_runtime": 146.1502, | |
| "eval_samples_per_second": 33.137, | |
| "eval_steps_per_second": 4.146, | |
| "eval_wer": 0.27039455009934715, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 44.11, | |
| "learning_rate": 9.964615384615385e-06, | |
| "loss": 1.0537, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 44.38, | |
| "learning_rate": 9.503076923076922e-06, | |
| "loss": 1.049, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 44.66, | |
| "learning_rate": 9.04153846153846e-06, | |
| "loss": 1.0517, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 44.93, | |
| "learning_rate": 8.58e-06, | |
| "loss": 1.036, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "learning_rate": 8.118461538461538e-06, | |
| "loss": 1.0786, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "eval_loss": 0.31082117557525635, | |
| "eval_runtime": 146.4144, | |
| "eval_samples_per_second": 33.077, | |
| "eval_steps_per_second": 4.139, | |
| "eval_wer": 0.270621629293216, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 45.48, | |
| "learning_rate": 7.656923076923076e-06, | |
| "loss": 1.0544, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 45.75, | |
| "learning_rate": 7.195384615384615e-06, | |
| "loss": 1.0702, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 46.03, | |
| "learning_rate": 6.733846153846153e-06, | |
| "loss": 1.035, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 46.3, | |
| "learning_rate": 6.272307692307692e-06, | |
| "loss": 1.0465, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 46.57, | |
| "learning_rate": 5.81076923076923e-06, | |
| "loss": 1.0546, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 46.57, | |
| "eval_loss": 0.31021571159362793, | |
| "eval_runtime": 145.9532, | |
| "eval_samples_per_second": 33.182, | |
| "eval_steps_per_second": 4.152, | |
| "eval_wer": 0.267726369571388, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 46.85, | |
| "learning_rate": 5.349230769230769e-06, | |
| "loss": 1.0473, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 47.12, | |
| "learning_rate": 4.887692307692307e-06, | |
| "loss": 1.0422, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 47.4, | |
| "learning_rate": 4.4261538461538455e-06, | |
| "loss": 1.0418, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 47.67, | |
| "learning_rate": 3.964615384615384e-06, | |
| "loss": 1.0509, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 47.94, | |
| "learning_rate": 3.5030769230769227e-06, | |
| "loss": 1.0425, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 47.94, | |
| "eval_loss": 0.3115251362323761, | |
| "eval_runtime": 146.7424, | |
| "eval_samples_per_second": 33.003, | |
| "eval_steps_per_second": 4.13, | |
| "eval_wer": 0.26792506386602327, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 48.22, | |
| "learning_rate": 3.0415384615384615e-06, | |
| "loss": 1.0463, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 48.49, | |
| "learning_rate": 2.58e-06, | |
| "loss": 1.0443, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 48.77, | |
| "learning_rate": 2.1230769230769228e-06, | |
| "loss": 1.0326, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 49.04, | |
| "learning_rate": 1.6615384615384613e-06, | |
| "loss": 1.0675, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 49.31, | |
| "learning_rate": 1.2e-06, | |
| "loss": 1.0398, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 49.31, | |
| "eval_loss": 0.31313779950141907, | |
| "eval_runtime": 145.9419, | |
| "eval_samples_per_second": 33.184, | |
| "eval_steps_per_second": 4.152, | |
| "eval_wer": 0.2665909736020437, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 49.59, | |
| "learning_rate": 7.384615384615384e-07, | |
| "loss": 1.017, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 49.86, | |
| "learning_rate": 2.769230769230769e-07, | |
| "loss": 1.0519, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 18250, | |
| "total_flos": 5.432980623924042e+19, | |
| "train_loss": 1.4442799015829009, | |
| "train_runtime": 29714.5027, | |
| "train_samples_per_second": 19.671, | |
| "train_steps_per_second": 0.614 | |
| } | |
| ], | |
| "max_steps": 18250, | |
| "num_train_epochs": 50, | |
| "total_flos": 5.432980623924042e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |