| { | |
| "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": 12.2128, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 7.4625e-06, | |
| "loss": 6.1116, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.1212499999999998e-05, | |
| "loss": 4.1396, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.49625e-05, | |
| "loss": 3.6725, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.8712499999999997e-05, | |
| "loss": 3.3224, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 3.335383415222168, | |
| "eval_runtime": 191.0801, | |
| "eval_samples_per_second": 25.345, | |
| "eval_steps_per_second": 3.171, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 2.2462499999999997e-05, | |
| "loss": 3.1394, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 2.6212499999999997e-05, | |
| "loss": 3.0669, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.99625e-05, | |
| "loss": 3.0401, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 3.37125e-05, | |
| "loss": 2.9719, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 3.7462499999999996e-05, | |
| "loss": 2.9318, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_loss": 2.93613600730896, | |
| "eval_runtime": 191.7956, | |
| "eval_samples_per_second": 25.251, | |
| "eval_steps_per_second": 3.16, | |
| "eval_wer": 1.0000283848992335, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.12125e-05, | |
| "loss": 2.9182, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 4.4962499999999995e-05, | |
| "loss": 2.8629, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 4.871249999999999e-05, | |
| "loss": 2.7891, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 5.2462499999999994e-05, | |
| "loss": 2.4819, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 5.62125e-05, | |
| "loss": 2.1371, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "eval_loss": 1.1156758069992065, | |
| "eval_runtime": 192.2259, | |
| "eval_samples_per_second": 25.194, | |
| "eval_steps_per_second": 3.153, | |
| "eval_wer": 0.8358501277320466, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 5.9962499999999994e-05, | |
| "loss": 1.9602, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 6.37125e-05, | |
| "loss": 1.8562, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 6.746249999999999e-05, | |
| "loss": 1.7867, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 7.121249999999999e-05, | |
| "loss": 1.7286, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 7.49625e-05, | |
| "loss": 1.6883, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "eval_loss": 0.6002867221832275, | |
| "eval_runtime": 191.4975, | |
| "eval_samples_per_second": 25.29, | |
| "eval_steps_per_second": 3.165, | |
| "eval_wer": 0.6314220834516038, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 7.454307692307691e-05, | |
| "loss": 1.6813, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 7.408153846153846e-05, | |
| "loss": 1.6522, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 7.361999999999999e-05, | |
| "loss": 1.6091, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 7.316307692307692e-05, | |
| "loss": 1.6031, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 7.270153846153845e-05, | |
| "loss": 1.5812, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_loss": 0.4746354818344116, | |
| "eval_runtime": 192.4568, | |
| "eval_samples_per_second": 25.164, | |
| "eval_steps_per_second": 3.149, | |
| "eval_wer": 0.4725234175418677, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 7.223999999999999e-05, | |
| "loss": 1.5692, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 7.177846153846153e-05, | |
| "loss": 1.5295, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 7.131692307692308e-05, | |
| "loss": 1.538, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 7.085538461538461e-05, | |
| "loss": 1.5169, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 7.039846153846153e-05, | |
| "loss": 1.5145, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "eval_loss": 0.437589168548584, | |
| "eval_runtime": 191.0945, | |
| "eval_samples_per_second": 25.343, | |
| "eval_steps_per_second": 3.171, | |
| "eval_wer": 0.4736304286119784, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 6.993692307692307e-05, | |
| "loss": 1.4912, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 6.94753846153846e-05, | |
| "loss": 1.4809, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 6.901384615384615e-05, | |
| "loss": 1.4801, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 6.855692307692307e-05, | |
| "loss": 1.4658, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 6.809538461538462e-05, | |
| "loss": 1.4763, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "eval_loss": 0.40057530999183655, | |
| "eval_runtime": 190.9848, | |
| "eval_samples_per_second": 25.358, | |
| "eval_steps_per_second": 3.173, | |
| "eval_wer": 0.38626170877093385, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 6.763384615384615e-05, | |
| "loss": 1.4436, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 6.717230769230768e-05, | |
| "loss": 1.4412, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 6.671076923076923e-05, | |
| "loss": 1.4332, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 6.624923076923076e-05, | |
| "loss": 1.4142, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 6.57876923076923e-05, | |
| "loss": 1.4215, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_loss": 0.3783426582813263, | |
| "eval_runtime": 191.1153, | |
| "eval_samples_per_second": 25.341, | |
| "eval_steps_per_second": 3.171, | |
| "eval_wer": 0.3628725518024411, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "learning_rate": 6.532615384615384e-05, | |
| "loss": 1.3928, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 6.486461538461538e-05, | |
| "loss": 1.408, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 6.440307692307692e-05, | |
| "loss": 1.3892, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 6.394153846153845e-05, | |
| "loss": 1.4071, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 12.33, | |
| "learning_rate": 6.348e-05, | |
| "loss": 1.3638, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.33, | |
| "eval_loss": 0.35546526312828064, | |
| "eval_runtime": 191.7841, | |
| "eval_samples_per_second": 25.252, | |
| "eval_steps_per_second": 3.16, | |
| "eval_wer": 0.3425205790519444, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 6.301846153846153e-05, | |
| "loss": 1.3788, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 12.88, | |
| "learning_rate": 6.255692307692307e-05, | |
| "loss": 1.3885, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 13.15, | |
| "learning_rate": 6.20953846153846e-05, | |
| "loss": 1.3793, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 13.42, | |
| "learning_rate": 6.163384615384614e-05, | |
| "loss": 1.3525, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "learning_rate": 6.117230769230768e-05, | |
| "loss": 1.3561, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "eval_loss": 0.33404484391212463, | |
| "eval_runtime": 191.5334, | |
| "eval_samples_per_second": 25.285, | |
| "eval_steps_per_second": 3.164, | |
| "eval_wer": 0.3227646891853534, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.97, | |
| "learning_rate": 6.0710769230769224e-05, | |
| "loss": 1.3548, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 6.024923076923076e-05, | |
| "loss": 1.3317, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 14.52, | |
| "learning_rate": 5.97876923076923e-05, | |
| "loss": 1.3356, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 5.9326153846153835e-05, | |
| "loss": 1.3564, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "learning_rate": 5.886461538461538e-05, | |
| "loss": 1.3406, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "eval_loss": 0.3372970223426819, | |
| "eval_runtime": 191.1166, | |
| "eval_samples_per_second": 25.341, | |
| "eval_steps_per_second": 3.171, | |
| "eval_wer": 0.32949191030371844, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.34, | |
| "learning_rate": 5.840307692307692e-05, | |
| "loss": 1.3158, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 5.794153846153845e-05, | |
| "loss": 1.3406, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 15.89, | |
| "learning_rate": 5.747999999999999e-05, | |
| "loss": 1.3312, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 16.16, | |
| "learning_rate": 5.701846153846153e-05, | |
| "loss": 1.3263, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "learning_rate": 5.655692307692308e-05, | |
| "loss": 1.3055, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "eval_loss": 0.3432324528694153, | |
| "eval_runtime": 190.6681, | |
| "eval_samples_per_second": 25.4, | |
| "eval_steps_per_second": 3.178, | |
| "eval_wer": 0.3210332103321033, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.71, | |
| "learning_rate": 5.609538461538461e-05, | |
| "loss": 1.3106, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 5.563384615384615e-05, | |
| "loss": 1.299, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 17.26, | |
| "learning_rate": 5.517230769230769e-05, | |
| "loss": 1.2942, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 5.471076923076922e-05, | |
| "loss": 1.2824, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 5.4249230769230766e-05, | |
| "loss": 1.3048, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "eval_loss": 0.3281940519809723, | |
| "eval_runtime": 190.4965, | |
| "eval_samples_per_second": 25.423, | |
| "eval_steps_per_second": 3.181, | |
| "eval_wer": 0.3117797331819472, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "learning_rate": 5.3787692307692306e-05, | |
| "loss": 1.2909, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 18.36, | |
| "learning_rate": 5.332615384615384e-05, | |
| "loss": 1.2934, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 18.63, | |
| "learning_rate": 5.286461538461538e-05, | |
| "loss": 1.2826, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 5.240307692307692e-05, | |
| "loss": 1.2683, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 5.194153846153846e-05, | |
| "loss": 1.2863, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "eval_loss": 0.32255104184150696, | |
| "eval_runtime": 191.9877, | |
| "eval_samples_per_second": 25.226, | |
| "eval_steps_per_second": 3.156, | |
| "eval_wer": 0.3018166335509509, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "learning_rate": 5.1479999999999995e-05, | |
| "loss": 1.2779, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 5.1018461538461534e-05, | |
| "loss": 1.2452, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 5.055692307692307e-05, | |
| "loss": 1.2789, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 20.27, | |
| "learning_rate": 5.009538461538461e-05, | |
| "loss": 1.2589, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 20.55, | |
| "learning_rate": 4.963846153846154e-05, | |
| "loss": 1.2389, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.55, | |
| "eval_loss": 0.3049624264240265, | |
| "eval_runtime": 191.4537, | |
| "eval_samples_per_second": 25.296, | |
| "eval_steps_per_second": 3.165, | |
| "eval_wer": 0.2985807550383196, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.82, | |
| "learning_rate": 4.917692307692307e-05, | |
| "loss": 1.2441, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 21.1, | |
| "learning_rate": 4.8719999999999994e-05, | |
| "loss": 1.2391, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 21.37, | |
| "learning_rate": 4.825846153846154e-05, | |
| "loss": 1.2227, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 21.64, | |
| "learning_rate": 4.779692307692307e-05, | |
| "loss": 1.2377, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "learning_rate": 4.734e-05, | |
| "loss": 1.2361, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "eval_loss": 0.30481210350990295, | |
| "eval_runtime": 190.3668, | |
| "eval_samples_per_second": 25.44, | |
| "eval_steps_per_second": 3.183, | |
| "eval_wer": 0.29804144195288107, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 22.19, | |
| "learning_rate": 4.687846153846153e-05, | |
| "loss": 1.2605, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 22.47, | |
| "learning_rate": 4.6416923076923076e-05, | |
| "loss": 1.2185, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 22.74, | |
| "learning_rate": 4.5955384615384615e-05, | |
| "loss": 1.2321, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 23.01, | |
| "learning_rate": 4.549384615384615e-05, | |
| "loss": 1.2674, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 23.29, | |
| "learning_rate": 4.503230769230769e-05, | |
| "loss": 1.2263, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 23.29, | |
| "eval_loss": 0.3011174499988556, | |
| "eval_runtime": 190.5, | |
| "eval_samples_per_second": 25.423, | |
| "eval_steps_per_second": 3.181, | |
| "eval_wer": 0.2977008231620778, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 23.56, | |
| "learning_rate": 4.4570769230769226e-05, | |
| "loss": 1.2177, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 23.83, | |
| "learning_rate": 4.410923076923077e-05, | |
| "loss": 1.2375, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 24.11, | |
| "learning_rate": 4.3647692307692304e-05, | |
| "loss": 1.2119, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 24.38, | |
| "learning_rate": 4.3186153846153844e-05, | |
| "loss": 1.221, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 24.66, | |
| "learning_rate": 4.2724615384615376e-05, | |
| "loss": 1.2225, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.66, | |
| "eval_loss": 0.3017021119594574, | |
| "eval_runtime": 191.4744, | |
| "eval_samples_per_second": 25.293, | |
| "eval_steps_per_second": 3.165, | |
| "eval_wer": 0.2958841896111269, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.93, | |
| "learning_rate": 4.2263076923076915e-05, | |
| "loss": 1.2195, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 25.21, | |
| "learning_rate": 4.180153846153846e-05, | |
| "loss": 1.2156, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 25.48, | |
| "learning_rate": 4.134e-05, | |
| "loss": 1.1914, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 25.75, | |
| "learning_rate": 4.087846153846153e-05, | |
| "loss": 1.1992, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 26.03, | |
| "learning_rate": 4.041692307692307e-05, | |
| "loss": 1.2044, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 26.03, | |
| "eval_loss": 0.29767629504203796, | |
| "eval_runtime": 190.7997, | |
| "eval_samples_per_second": 25.383, | |
| "eval_steps_per_second": 3.176, | |
| "eval_wer": 0.2782287822878229, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 26.3, | |
| "learning_rate": 3.995538461538461e-05, | |
| "loss": 1.1957, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 26.57, | |
| "learning_rate": 3.949384615384615e-05, | |
| "loss": 1.184, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 26.85, | |
| "learning_rate": 3.903230769230769e-05, | |
| "loss": 1.1873, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "learning_rate": 3.857076923076923e-05, | |
| "loss": 1.1762, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "learning_rate": 3.810923076923076e-05, | |
| "loss": 1.2017, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "eval_loss": 0.29657357931137085, | |
| "eval_runtime": 191.6103, | |
| "eval_samples_per_second": 25.275, | |
| "eval_steps_per_second": 3.163, | |
| "eval_wer": 0.27814362759012207, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 27.67, | |
| "learning_rate": 3.76476923076923e-05, | |
| "loss": 1.1837, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 27.94, | |
| "learning_rate": 3.718615384615384e-05, | |
| "loss": 1.1948, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 28.22, | |
| "learning_rate": 3.6724615384615386e-05, | |
| "loss": 1.1852, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 28.49, | |
| "learning_rate": 3.626307692307692e-05, | |
| "loss": 1.1681, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "learning_rate": 3.580153846153846e-05, | |
| "loss": 1.1912, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "eval_loss": 0.2999463379383087, | |
| "eval_runtime": 192.5233, | |
| "eval_samples_per_second": 25.155, | |
| "eval_steps_per_second": 3.148, | |
| "eval_wer": 0.2785977859778598, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 29.04, | |
| "learning_rate": 3.534e-05, | |
| "loss": 1.1658, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 29.31, | |
| "learning_rate": 3.4878461538461536e-05, | |
| "loss": 1.1691, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 29.59, | |
| "learning_rate": 3.4416923076923076e-05, | |
| "loss": 1.1531, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 29.86, | |
| "learning_rate": 3.3955384615384615e-05, | |
| "loss": 1.1569, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 30.14, | |
| "learning_rate": 3.349384615384615e-05, | |
| "loss": 1.1658, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 30.14, | |
| "eval_loss": 0.29912158846855164, | |
| "eval_runtime": 192.0401, | |
| "eval_samples_per_second": 25.219, | |
| "eval_steps_per_second": 3.156, | |
| "eval_wer": 0.2757309111552654, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 30.41, | |
| "learning_rate": 3.303230769230769e-05, | |
| "loss": 1.1565, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 30.68, | |
| "learning_rate": 3.2570769230769226e-05, | |
| "loss": 1.1556, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 30.96, | |
| "learning_rate": 3.2109230769230765e-05, | |
| "loss": 1.1538, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 31.23, | |
| "learning_rate": 3.1647692307692304e-05, | |
| "loss": 1.1531, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 31.51, | |
| "learning_rate": 3.118615384615384e-05, | |
| "loss": 1.148, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 31.51, | |
| "eval_loss": 0.2915021479129791, | |
| "eval_runtime": 191.8689, | |
| "eval_samples_per_second": 25.241, | |
| "eval_steps_per_second": 3.158, | |
| "eval_wer": 0.268379222253761, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 31.78, | |
| "learning_rate": 3.072461538461538e-05, | |
| "loss": 1.1414, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 32.05, | |
| "learning_rate": 3.026307692307692e-05, | |
| "loss": 1.1364, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 32.33, | |
| "learning_rate": 2.980153846153846e-05, | |
| "loss": 1.1531, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 32.6, | |
| "learning_rate": 2.9339999999999997e-05, | |
| "loss": 1.1331, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 32.88, | |
| "learning_rate": 2.8878461538461536e-05, | |
| "loss": 1.1423, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 32.88, | |
| "eval_loss": 0.29132989048957825, | |
| "eval_runtime": 191.3127, | |
| "eval_samples_per_second": 25.315, | |
| "eval_steps_per_second": 3.168, | |
| "eval_wer": 0.2642634118648879, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 33.15, | |
| "learning_rate": 2.8416923076923075e-05, | |
| "loss": 1.1527, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 33.42, | |
| "learning_rate": 2.795538461538461e-05, | |
| "loss": 1.1155, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 33.7, | |
| "learning_rate": 2.7498461538461536e-05, | |
| "loss": 1.1082, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 33.97, | |
| "learning_rate": 2.703692307692307e-05, | |
| "loss": 1.1172, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 34.25, | |
| "learning_rate": 2.6575384615384614e-05, | |
| "loss": 1.123, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 34.25, | |
| "eval_loss": 0.2776812016963959, | |
| "eval_runtime": 191.5218, | |
| "eval_samples_per_second": 25.287, | |
| "eval_steps_per_second": 3.164, | |
| "eval_wer": 0.26304286119784276, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 34.52, | |
| "learning_rate": 2.611384615384615e-05, | |
| "loss": 1.1189, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 34.79, | |
| "learning_rate": 2.5652307692307693e-05, | |
| "loss": 1.1115, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 35.07, | |
| "learning_rate": 2.519076923076923e-05, | |
| "loss": 1.1182, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 35.34, | |
| "learning_rate": 2.4729230769230764e-05, | |
| "loss": 1.1172, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 35.62, | |
| "learning_rate": 2.4267692307692307e-05, | |
| "loss": 1.1297, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 35.62, | |
| "eval_loss": 0.28734585642814636, | |
| "eval_runtime": 191.3152, | |
| "eval_samples_per_second": 25.314, | |
| "eval_steps_per_second": 3.168, | |
| "eval_wer": 0.26460403065569116, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 35.89, | |
| "learning_rate": 2.3806153846153843e-05, | |
| "loss": 1.1112, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 36.16, | |
| "learning_rate": 2.3349230769230767e-05, | |
| "loss": 1.1157, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 36.44, | |
| "learning_rate": 2.2887692307692303e-05, | |
| "loss": 1.0938, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 36.71, | |
| "learning_rate": 2.2426153846153846e-05, | |
| "loss": 1.1007, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 36.98, | |
| "learning_rate": 2.1964615384615382e-05, | |
| "loss": 1.0987, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 36.98, | |
| "eval_loss": 0.28285086154937744, | |
| "eval_runtime": 190.4602, | |
| "eval_samples_per_second": 25.428, | |
| "eval_steps_per_second": 3.182, | |
| "eval_wer": 0.2619358501277321, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 37.26, | |
| "learning_rate": 2.1503076923076924e-05, | |
| "loss": 1.0967, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 37.53, | |
| "learning_rate": 2.104153846153846e-05, | |
| "loss": 1.0922, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 37.81, | |
| "learning_rate": 2.0579999999999996e-05, | |
| "loss": 1.0885, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 38.08, | |
| "learning_rate": 2.011846153846154e-05, | |
| "loss": 1.0979, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 38.36, | |
| "learning_rate": 1.9656923076923075e-05, | |
| "loss": 1.0873, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 38.36, | |
| "eval_loss": 0.2864430546760559, | |
| "eval_runtime": 191.1374, | |
| "eval_samples_per_second": 25.338, | |
| "eval_steps_per_second": 3.17, | |
| "eval_wer": 0.26080045415838776, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 38.63, | |
| "learning_rate": 1.9195384615384614e-05, | |
| "loss": 1.0868, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 38.9, | |
| "learning_rate": 1.8733846153846153e-05, | |
| "loss": 1.0831, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 39.18, | |
| "learning_rate": 1.8272307692307692e-05, | |
| "loss": 1.0907, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 39.45, | |
| "learning_rate": 1.781076923076923e-05, | |
| "loss": 1.093, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 39.73, | |
| "learning_rate": 1.7349230769230767e-05, | |
| "loss": 1.0848, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 39.73, | |
| "eval_loss": 0.2827305793762207, | |
| "eval_runtime": 193.1681, | |
| "eval_samples_per_second": 25.071, | |
| "eval_steps_per_second": 3.137, | |
| "eval_wer": 0.2576781152426909, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 1.6887692307692307e-05, | |
| "loss": 1.0913, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 40.27, | |
| "learning_rate": 1.6426153846153846e-05, | |
| "loss": 1.0764, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 40.55, | |
| "learning_rate": 1.5964615384615385e-05, | |
| "loss": 1.0668, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 40.82, | |
| "learning_rate": 1.550307692307692e-05, | |
| "loss": 1.0845, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 41.1, | |
| "learning_rate": 1.5041538461538462e-05, | |
| "loss": 1.0628, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 41.1, | |
| "eval_loss": 0.28957271575927734, | |
| "eval_runtime": 192.7832, | |
| "eval_samples_per_second": 25.121, | |
| "eval_steps_per_second": 3.143, | |
| "eval_wer": 0.258103888731195, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 41.37, | |
| "learning_rate": 1.4579999999999998e-05, | |
| "loss": 1.0746, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 41.64, | |
| "learning_rate": 1.4118461538461537e-05, | |
| "loss": 1.0889, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 41.92, | |
| "learning_rate": 1.3656923076923076e-05, | |
| "loss": 1.0669, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 42.19, | |
| "learning_rate": 1.3195384615384615e-05, | |
| "loss": 1.069, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 42.47, | |
| "learning_rate": 1.2733846153846153e-05, | |
| "loss": 1.0815, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 42.47, | |
| "eval_loss": 0.28144747018814087, | |
| "eval_runtime": 192.799, | |
| "eval_samples_per_second": 25.119, | |
| "eval_steps_per_second": 3.143, | |
| "eval_wer": 0.25608856088560883, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 42.74, | |
| "learning_rate": 1.227230769230769e-05, | |
| "loss": 1.0468, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 1.1815384615384615e-05, | |
| "loss": 1.073, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 43.29, | |
| "learning_rate": 1.1353846153846154e-05, | |
| "loss": 1.05, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 43.56, | |
| "learning_rate": 1.089230769230769e-05, | |
| "loss": 1.0562, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 43.83, | |
| "learning_rate": 1.043076923076923e-05, | |
| "loss": 1.0587, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 43.83, | |
| "eval_loss": 0.27383875846862793, | |
| "eval_runtime": 191.944, | |
| "eval_samples_per_second": 25.231, | |
| "eval_steps_per_second": 3.157, | |
| "eval_wer": 0.25421515753619073, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 44.11, | |
| "learning_rate": 9.969230769230769e-06, | |
| "loss": 1.0479, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 44.38, | |
| "learning_rate": 9.507692307692308e-06, | |
| "loss": 1.0464, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 44.66, | |
| "learning_rate": 9.046153846153845e-06, | |
| "loss": 1.0515, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 44.93, | |
| "learning_rate": 8.584615384615385e-06, | |
| "loss": 1.0456, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "learning_rate": 8.123076923076922e-06, | |
| "loss": 1.0709, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 45.21, | |
| "eval_loss": 0.27852663397789, | |
| "eval_runtime": 191.9755, | |
| "eval_samples_per_second": 25.227, | |
| "eval_steps_per_second": 3.157, | |
| "eval_wer": 0.25784842463809254, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 45.48, | |
| "learning_rate": 7.661538461538461e-06, | |
| "loss": 1.0616, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 45.75, | |
| "learning_rate": 7.2e-06, | |
| "loss": 1.0707, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 46.03, | |
| "learning_rate": 6.738461538461537e-06, | |
| "loss": 1.037, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 46.3, | |
| "learning_rate": 6.2769230769230764e-06, | |
| "loss": 1.0475, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 46.57, | |
| "learning_rate": 5.815384615384615e-06, | |
| "loss": 1.0512, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 46.57, | |
| "eval_loss": 0.27928975224494934, | |
| "eval_runtime": 191.7381, | |
| "eval_samples_per_second": 25.258, | |
| "eval_steps_per_second": 3.161, | |
| "eval_wer": 0.2539029236446211, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 46.85, | |
| "learning_rate": 5.353846153846153e-06, | |
| "loss": 1.0454, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 47.12, | |
| "learning_rate": 4.892307692307692e-06, | |
| "loss": 1.0505, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 47.4, | |
| "learning_rate": 4.43076923076923e-06, | |
| "loss": 1.0351, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 47.67, | |
| "learning_rate": 3.969230769230768e-06, | |
| "loss": 1.0591, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 47.94, | |
| "learning_rate": 3.5076923076923076e-06, | |
| "loss": 1.0396, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 47.94, | |
| "eval_loss": 0.27876144647598267, | |
| "eval_runtime": 191.8086, | |
| "eval_samples_per_second": 25.249, | |
| "eval_steps_per_second": 3.159, | |
| "eval_wer": 0.25251206358217426, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 48.22, | |
| "learning_rate": 3.046153846153846e-06, | |
| "loss": 1.0482, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 48.49, | |
| "learning_rate": 2.5846153846153844e-06, | |
| "loss": 1.0433, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 48.77, | |
| "learning_rate": 2.1230769230769228e-06, | |
| "loss": 1.036, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 49.04, | |
| "learning_rate": 1.6615384615384613e-06, | |
| "loss": 1.0739, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 49.31, | |
| "learning_rate": 1.2046153846153845e-06, | |
| "loss": 1.0481, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 49.31, | |
| "eval_loss": 0.27769750356674194, | |
| "eval_runtime": 191.0585, | |
| "eval_samples_per_second": 25.348, | |
| "eval_steps_per_second": 3.172, | |
| "eval_wer": 0.2534487652568833, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 49.59, | |
| "learning_rate": 7.43076923076923e-07, | |
| "loss": 1.0195, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 49.86, | |
| "learning_rate": 2.8153846153846154e-07, | |
| "loss": 1.0473, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 18250, | |
| "total_flos": 5.432980623924042e+19, | |
| "train_loss": 1.449225697974636, | |
| "train_runtime": 31196.8756, | |
| "train_samples_per_second": 18.736, | |
| "train_steps_per_second": 0.585 | |
| } | |
| ], | |
| "max_steps": 18250, | |
| "num_train_epochs": 50, | |
| "total_flos": 5.432980623924042e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |