| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.999581414817916, | |
| "global_step": 5970, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.9000000000000005e-05, | |
| "loss": 5.3488, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 3.2572, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.000149, | |
| "loss": 2.4392, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.000199, | |
| "loss": 2.2566, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.000249, | |
| "loss": 2.2416, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 1.2866647243499756, | |
| "eval_runtime": 434.4225, | |
| "eval_samples_per_second": 23.912, | |
| "eval_steps_per_second": 2.99, | |
| "eval_wer": 0.8874521338587047, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.000299, | |
| "loss": 2.2596, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00034899999999999997, | |
| "loss": 2.2575, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00039900000000000005, | |
| "loss": 2.2978, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.000449, | |
| "loss": 2.2998, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.000499, | |
| "loss": 2.3089, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 1.8336485624313354, | |
| "eval_runtime": 430.7741, | |
| "eval_samples_per_second": 24.115, | |
| "eval_steps_per_second": 3.016, | |
| "eval_wer": 0.9547514660451005, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.000549, | |
| "loss": 2.3156, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.000599, | |
| "loss": 2.3298, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0006490000000000001, | |
| "loss": 2.3174, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.000699, | |
| "loss": 2.349, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.000749, | |
| "loss": 2.3614, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_loss": 1.5936506986618042, | |
| "eval_runtime": 430.1389, | |
| "eval_samples_per_second": 24.15, | |
| "eval_steps_per_second": 3.02, | |
| "eval_wer": 0.9468893945279983, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.000799, | |
| "loss": 2.3597, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.000849, | |
| "loss": 2.397, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.0008990000000000001, | |
| "loss": 2.4454, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 0.000949, | |
| "loss": 2.4806, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.000999, | |
| "loss": 2.5234, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_loss": 1.9764641523361206, | |
| "eval_runtime": 434.6883, | |
| "eval_samples_per_second": 23.898, | |
| "eval_steps_per_second": 2.988, | |
| "eval_wer": 0.9866992248922434, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 0.0009753148614609572, | |
| "loss": 2.5559, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.0009501259445843828, | |
| "loss": 2.5543, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.0009249370277078086, | |
| "loss": 2.5819, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 0.0008997481108312343, | |
| "loss": 2.5837, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 0.00087455919395466, | |
| "loss": 2.5373, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_loss": 1.9062319993972778, | |
| "eval_runtime": 430.3022, | |
| "eval_samples_per_second": 24.141, | |
| "eval_steps_per_second": 3.019, | |
| "eval_wer": 0.9916014577205542, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 0.0008493702770780856, | |
| "loss": 2.5617, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.0008241813602015113, | |
| "loss": 2.5553, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 0.0007989924433249371, | |
| "loss": 2.549, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 0.0007738035264483628, | |
| "loss": 2.5636, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 0.0007486146095717884, | |
| "loss": 2.5703, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_loss": 1.977164387702942, | |
| "eval_runtime": 431.1212, | |
| "eval_samples_per_second": 24.095, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 0.9914719647779197, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.0007234256926952141, | |
| "loss": 2.5526, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 0.0006982367758186398, | |
| "loss": 2.5277, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.0006730478589420656, | |
| "loss": 2.4969, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 0.0006478589420654912, | |
| "loss": 2.4849, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.0006226700251889169, | |
| "loss": 2.4656, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "eval_loss": 1.8083465099334717, | |
| "eval_runtime": 432.7157, | |
| "eval_samples_per_second": 24.007, | |
| "eval_steps_per_second": 3.002, | |
| "eval_wer": 0.9829069315722293, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 0.0005974811083123426, | |
| "loss": 2.4858, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 0.0005722921914357682, | |
| "loss": 2.4552, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 0.0005471032745591939, | |
| "loss": 2.4302, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 0.0005219143576826196, | |
| "loss": 2.4397, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 0.0004967254408060454, | |
| "loss": 2.4339, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "eval_loss": 1.754757046699524, | |
| "eval_runtime": 442.6863, | |
| "eval_samples_per_second": 23.466, | |
| "eval_steps_per_second": 2.934, | |
| "eval_wer": 0.9752483489649814, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 0.00047153652392947104, | |
| "loss": 2.4069, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 0.0004463476070528967, | |
| "loss": 2.3863, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.00042115869017632243, | |
| "loss": 2.3614, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.0003959697732997481, | |
| "loss": 2.3534, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 0.00037078085642317383, | |
| "loss": 2.344, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "eval_loss": 1.6146422624588013, | |
| "eval_runtime": 432.0313, | |
| "eval_samples_per_second": 24.045, | |
| "eval_steps_per_second": 3.007, | |
| "eval_wer": 0.9638344710213294, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 0.0003455919395465995, | |
| "loss": 2.3322, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 0.0003204030226700252, | |
| "loss": 2.315, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 0.00029521410579345085, | |
| "loss": 2.3035, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 0.0002700251889168766, | |
| "loss": 2.2715, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 0.00024483627204030224, | |
| "loss": 2.2677, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "eval_loss": 1.5104962587356567, | |
| "eval_runtime": 431.1839, | |
| "eval_samples_per_second": 24.092, | |
| "eval_steps_per_second": 3.013, | |
| "eval_wer": 0.9499232291840095, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.54, | |
| "learning_rate": 0.00021964735516372797, | |
| "loss": 2.266, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 0.00019445843828715364, | |
| "loss": 2.2473, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 0.00016926952141057937, | |
| "loss": 2.2419, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 0.00014408060453400504, | |
| "loss": 2.2305, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 0.00011889168765743074, | |
| "loss": 2.2074, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "eval_loss": 1.4190884828567505, | |
| "eval_runtime": 440.4999, | |
| "eval_samples_per_second": 23.582, | |
| "eval_steps_per_second": 2.949, | |
| "eval_wer": 0.9356790054942006, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 9.370277078085642e-05, | |
| "loss": 2.1936, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 6.851385390428212e-05, | |
| "loss": 2.1796, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "learning_rate": 4.332493702770781e-05, | |
| "loss": 2.1748, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 1.8136020151133502e-05, | |
| "loss": 2.1846, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 5970, | |
| "total_flos": 1.0051413716540667e+20, | |
| "train_loss": 2.4458747799871756, | |
| "train_runtime": 25997.2696, | |
| "train_samples_per_second": 14.697, | |
| "train_steps_per_second": 0.23 | |
| } | |
| ], | |
| "max_steps": 5970, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.0051413716540667e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |