| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 137.09493670886076, | |
| "global_step": 4250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 3.5e-05, | |
| "loss": 7.2926, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.8496713638305664, | |
| "eval_runtime": 271.7106, | |
| "eval_samples_per_second": 26.745, | |
| "eval_steps_per_second": 3.345, | |
| "eval_wer": 1.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 7e-05, | |
| "loss": 3.417, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_cer": 0.9856759584948347, | |
| "eval_loss": 3.285226583480835, | |
| "eval_runtime": 275.8724, | |
| "eval_samples_per_second": 26.342, | |
| "eval_steps_per_second": 3.295, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "learning_rate": 6.578313253012048e-05, | |
| "loss": 2.0264, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 24.19, | |
| "eval_cer": 0.176804441300443, | |
| "eval_loss": 0.7098603248596191, | |
| "eval_runtime": 269.2535, | |
| "eval_samples_per_second": 26.989, | |
| "eval_steps_per_second": 3.376, | |
| "eval_wer": 0.7342138090806487, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 32.25, | |
| "learning_rate": 6.156626506024095e-05, | |
| "loss": 0.4018, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 32.25, | |
| "eval_cer": 0.15511300475620768, | |
| "eval_loss": 0.6187673211097717, | |
| "eval_runtime": 271.6934, | |
| "eval_samples_per_second": 26.747, | |
| "eval_steps_per_second": 3.346, | |
| "eval_wer": 0.6415460467694989, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "learning_rate": 5.734939759036144e-05, | |
| "loss": 0.2444, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 40.32, | |
| "eval_cer": 0.15995397008055237, | |
| "eval_loss": 0.6631603837013245, | |
| "eval_runtime": 276.2198, | |
| "eval_samples_per_second": 26.309, | |
| "eval_steps_per_second": 3.291, | |
| "eval_wer": 0.6361800289091737, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 48.38, | |
| "learning_rate": 5.313253012048192e-05, | |
| "loss": 0.1882, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 48.38, | |
| "eval_cer": 0.13876129966064343, | |
| "eval_loss": 0.6070172190666199, | |
| "eval_runtime": 272.4136, | |
| "eval_samples_per_second": 26.676, | |
| "eval_steps_per_second": 3.337, | |
| "eval_wer": 0.578262677464705, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 56.44, | |
| "learning_rate": 4.891566265060241e-05, | |
| "loss": 0.153, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 56.44, | |
| "eval_cer": 0.13767574986063888, | |
| "eval_loss": 0.6425250172615051, | |
| "eval_runtime": 272.6101, | |
| "eval_samples_per_second": 26.657, | |
| "eval_steps_per_second": 3.334, | |
| "eval_wer": 0.5720056234283112, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 64.51, | |
| "learning_rate": 4.469879518072288e-05, | |
| "loss": 0.1214, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 64.51, | |
| "eval_cer": 0.1337117001405021, | |
| "eval_loss": 0.6362873315811157, | |
| "eval_runtime": 271.8138, | |
| "eval_samples_per_second": 26.735, | |
| "eval_steps_per_second": 3.344, | |
| "eval_wer": 0.5546007167891016, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 72.57, | |
| "learning_rate": 4.048192771084337e-05, | |
| "loss": 0.1011, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 72.57, | |
| "eval_cer": 0.12238677519991394, | |
| "eval_loss": 0.6309632658958435, | |
| "eval_runtime": 276.7543, | |
| "eval_samples_per_second": 26.258, | |
| "eval_steps_per_second": 3.285, | |
| "eval_wer": 0.5221669999801992, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 80.63, | |
| "learning_rate": 3.6265060240963855e-05, | |
| "loss": 0.0879, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 80.63, | |
| "eval_cer": 0.12531743366899534, | |
| "eval_loss": 0.6352854371070862, | |
| "eval_runtime": 270.8217, | |
| "eval_samples_per_second": 26.833, | |
| "eval_steps_per_second": 3.356, | |
| "eval_wer": 0.5258301487040374, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 88.7, | |
| "learning_rate": 3.2048192771084335e-05, | |
| "loss": 0.0782, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 88.7, | |
| "eval_cer": 0.11265268600227542, | |
| "eval_loss": 0.607792854309082, | |
| "eval_runtime": 270.7843, | |
| "eval_samples_per_second": 26.837, | |
| "eval_steps_per_second": 3.357, | |
| "eval_wer": 0.4904263113082391, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 96.76, | |
| "learning_rate": 2.783132530120482e-05, | |
| "loss": 0.0709, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 96.76, | |
| "eval_cer": 0.11539100982210675, | |
| "eval_loss": 0.6464908123016357, | |
| "eval_runtime": 272.9311, | |
| "eval_samples_per_second": 26.626, | |
| "eval_steps_per_second": 3.331, | |
| "eval_wer": 0.49601013801160326, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 104.82, | |
| "learning_rate": 2.36144578313253e-05, | |
| "loss": 0.0661, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 104.82, | |
| "eval_cer": 0.11656783708277235, | |
| "eval_loss": 0.6621575951576233, | |
| "eval_runtime": 270.3992, | |
| "eval_samples_per_second": 26.875, | |
| "eval_steps_per_second": 3.362, | |
| "eval_wer": 0.494544878522068, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 112.89, | |
| "learning_rate": 1.9397590361445782e-05, | |
| "loss": 0.0616, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 112.89, | |
| "eval_cer": 0.11035119002989337, | |
| "eval_loss": 0.6440250277519226, | |
| "eval_runtime": 271.0109, | |
| "eval_samples_per_second": 26.814, | |
| "eval_steps_per_second": 3.354, | |
| "eval_wer": 0.47860523137239375, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 120.95, | |
| "learning_rate": 1.5180722891566264e-05, | |
| "loss": 0.0579, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 120.95, | |
| "eval_cer": 0.11444237621309375, | |
| "eval_loss": 0.6815317273139954, | |
| "eval_runtime": 269.1998, | |
| "eval_samples_per_second": 26.995, | |
| "eval_steps_per_second": 3.377, | |
| "eval_wer": 0.4887432429756648, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 129.03, | |
| "learning_rate": 1.0963855421686746e-05, | |
| "loss": 0.0549, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 129.03, | |
| "eval_cer": 0.11051744540466885, | |
| "eval_loss": 0.6602992415428162, | |
| "eval_runtime": 274.1737, | |
| "eval_samples_per_second": 26.505, | |
| "eval_steps_per_second": 3.315, | |
| "eval_wer": 0.47799140645110194, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 137.09, | |
| "learning_rate": 6.746987951807228e-06, | |
| "loss": 0.0527, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 137.09, | |
| "eval_cer": 0.10900158757583364, | |
| "eval_loss": 0.6652226448059082, | |
| "eval_runtime": 271.842, | |
| "eval_samples_per_second": 26.732, | |
| "eval_steps_per_second": 3.344, | |
| "eval_wer": 0.47486287943290495, | |
| "step": 4250 | |
| } | |
| ], | |
| "max_steps": 4650, | |
| "num_train_epochs": 150, | |
| "total_flos": 3.524250563411945e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |