| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.67427122940431, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 3.0458309650421143, | |
| "eval_runtime": 143.841, | |
| "eval_samples_per_second": 32.835, | |
| "eval_steps_per_second": 2.739, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 9.98e-05, | |
| "loss": 2.9181, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 1.1502786874771118, | |
| "eval_runtime": 143.5317, | |
| "eval_samples_per_second": 32.906, | |
| "eval_steps_per_second": 2.745, | |
| "eval_wer": 0.4722885713129023, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 0.8274601101875305, | |
| "eval_runtime": 142.4484, | |
| "eval_samples_per_second": 33.156, | |
| "eval_steps_per_second": 2.766, | |
| "eval_wer": 0.3499858305331768, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 8.89111111111111e-05, | |
| "loss": 0.7646, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 0.6990081667900085, | |
| "eval_runtime": 145.8955, | |
| "eval_samples_per_second": 32.372, | |
| "eval_steps_per_second": 2.701, | |
| "eval_wer": 0.2845228938099672, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "eval_loss": 0.5828294157981873, | |
| "eval_runtime": 147.7617, | |
| "eval_samples_per_second": 31.964, | |
| "eval_steps_per_second": 2.666, | |
| "eval_wer": 0.2508602890571232, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 7.780000000000001e-05, | |
| "loss": 0.5394, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "eval_loss": 0.5363064408302307, | |
| "eval_runtime": 147.8939, | |
| "eval_samples_per_second": 31.935, | |
| "eval_steps_per_second": 2.664, | |
| "eval_wer": 0.24873486903364236, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "eval_loss": 0.5467007756233215, | |
| "eval_runtime": 151.8112, | |
| "eval_samples_per_second": 31.111, | |
| "eval_steps_per_second": 2.595, | |
| "eval_wer": 0.21711671592243229, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 6.66888888888889e-05, | |
| "loss": 0.4558, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "eval_loss": 0.5289521813392639, | |
| "eval_runtime": 152.3052, | |
| "eval_samples_per_second": 31.01, | |
| "eval_steps_per_second": 2.587, | |
| "eval_wer": 0.20901987773774341, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "eval_loss": 0.4992470443248749, | |
| "eval_runtime": 148.0991, | |
| "eval_samples_per_second": 31.891, | |
| "eval_steps_per_second": 2.66, | |
| "eval_wer": 0.20464758511801143, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 5.558888888888889e-05, | |
| "loss": 0.3773, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "eval_loss": 0.49340707063674927, | |
| "eval_runtime": 155.8909, | |
| "eval_samples_per_second": 30.297, | |
| "eval_steps_per_second": 2.527, | |
| "eval_wer": 0.2051738796000162, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_loss": 0.47003018856048584, | |
| "eval_runtime": 146.5312, | |
| "eval_samples_per_second": 32.232, | |
| "eval_steps_per_second": 2.689, | |
| "eval_wer": 0.19827132504756892, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 4.447777777777778e-05, | |
| "loss": 0.3301, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "eval_loss": 0.4938233494758606, | |
| "eval_runtime": 154.4839, | |
| "eval_samples_per_second": 30.573, | |
| "eval_steps_per_second": 2.55, | |
| "eval_wer": 0.18738107768916237, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "eval_loss": 0.5363861322402954, | |
| "eval_runtime": 163.224, | |
| "eval_samples_per_second": 28.936, | |
| "eval_steps_per_second": 2.414, | |
| "eval_wer": 0.18934456094894944, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 3.336666666666667e-05, | |
| "loss": 0.2938, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "eval_loss": 0.5169841051101685, | |
| "eval_runtime": 164.5878, | |
| "eval_samples_per_second": 28.696, | |
| "eval_steps_per_second": 2.394, | |
| "eval_wer": 0.18300878506943039, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "eval_loss": 0.5407580733299255, | |
| "eval_runtime": 163.5687, | |
| "eval_samples_per_second": 28.875, | |
| "eval_steps_per_second": 2.409, | |
| "eval_wer": 0.18149062790980122, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 2.225555555555556e-05, | |
| "loss": 0.2674, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "eval_loss": 0.5580623149871826, | |
| "eval_runtime": 162.7994, | |
| "eval_samples_per_second": 29.011, | |
| "eval_steps_per_second": 2.42, | |
| "eval_wer": 0.173272337152342, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "eval_loss": 0.538887083530426, | |
| "eval_runtime": 163.5851, | |
| "eval_samples_per_second": 28.872, | |
| "eval_steps_per_second": 2.409, | |
| "eval_wer": 0.1719161167564066, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 1.1144444444444445e-05, | |
| "loss": 0.24, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "eval_loss": 0.5344488024711609, | |
| "eval_runtime": 163.7329, | |
| "eval_samples_per_second": 28.846, | |
| "eval_steps_per_second": 2.406, | |
| "eval_wer": 0.17141006436986356, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "eval_loss": 0.5502966046333313, | |
| "eval_runtime": 154.8226, | |
| "eval_samples_per_second": 30.506, | |
| "eval_steps_per_second": 2.545, | |
| "eval_wer": 0.16863689729160763, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 3.3333333333333334e-08, | |
| "loss": 0.211, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "eval_loss": 0.5595183372497559, | |
| "eval_runtime": 157.1471, | |
| "eval_samples_per_second": 30.055, | |
| "eval_steps_per_second": 2.507, | |
| "eval_wer": 0.16865713938706936, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "step": 10000, | |
| "total_flos": 2.7136568655380324e+19, | |
| "train_loss": 0.6397389862060547, | |
| "train_runtime": 13054.0758, | |
| "train_samples_per_second": 18.385, | |
| "train_steps_per_second": 0.766 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 13, | |
| "total_flos": 2.7136568655380324e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |