| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.99521531100478, | |
| "global_step": 5200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3.2333333333333334e-06, | |
| "loss": 13.5304, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 6.566666666666667e-06, | |
| "loss": 3.5412, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 9.9e-06, | |
| "loss": 3.1425, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 9.80204081632653e-06, | |
| "loss": 2.0204, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 9.59795918367347e-06, | |
| "loss": 1.5794, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "eval_loss": 0.9859886169433594, | |
| "eval_runtime": 65.6063, | |
| "eval_samples_per_second": 16.95, | |
| "eval_steps_per_second": 4.237, | |
| "eval_wer": 0.9582062780269058, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 9.39387755102041e-06, | |
| "loss": 1.4015, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "learning_rate": 9.189795918367347e-06, | |
| "loss": 1.2803, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 8.985714285714287e-06, | |
| "loss": 1.1969, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 8.781632653061225e-06, | |
| "loss": 1.1458, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 8.577551020408163e-06, | |
| "loss": 1.0985, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_loss": 0.5948446989059448, | |
| "eval_runtime": 63.7089, | |
| "eval_samples_per_second": 17.454, | |
| "eval_steps_per_second": 4.364, | |
| "eval_wer": 0.7533632286995515, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "learning_rate": 8.373469387755104e-06, | |
| "loss": 1.0639, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 8.169387755102042e-06, | |
| "loss": 1.0214, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 7.96530612244898e-06, | |
| "loss": 1.0047, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 7.76122448979592e-06, | |
| "loss": 0.971, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 28.84, | |
| "learning_rate": 7.557142857142857e-06, | |
| "loss": 0.9551, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 28.84, | |
| "eval_loss": 0.5892378091812134, | |
| "eval_runtime": 64.7699, | |
| "eval_samples_per_second": 17.168, | |
| "eval_steps_per_second": 4.292, | |
| "eval_wer": 0.7391928251121076, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 7.353061224489797e-06, | |
| "loss": 0.9387, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 32.69, | |
| "learning_rate": 7.148979591836735e-06, | |
| "loss": 0.9272, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 34.61, | |
| "learning_rate": 6.944897959183674e-06, | |
| "loss": 0.9192, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 36.54, | |
| "learning_rate": 6.740816326530613e-06, | |
| "loss": 0.892, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 6.5367346938775515e-06, | |
| "loss": 0.8699, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_loss": 0.5926958322525024, | |
| "eval_runtime": 64.9144, | |
| "eval_samples_per_second": 17.13, | |
| "eval_steps_per_second": 4.283, | |
| "eval_wer": 0.72, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 40.38, | |
| "learning_rate": 6.332653061224491e-06, | |
| "loss": 0.8779, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 42.31, | |
| "learning_rate": 6.128571428571429e-06, | |
| "loss": 0.8511, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 44.23, | |
| "learning_rate": 5.924489795918368e-06, | |
| "loss": 0.8532, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 5.720408163265306e-06, | |
| "loss": 0.8442, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 48.08, | |
| "learning_rate": 5.516326530612245e-06, | |
| "loss": 0.8372, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 48.08, | |
| "eval_loss": 0.5938696265220642, | |
| "eval_runtime": 64.5641, | |
| "eval_samples_per_second": 17.223, | |
| "eval_steps_per_second": 4.306, | |
| "eval_wer": 0.73847533632287, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 5.312244897959185e-06, | |
| "loss": 0.8112, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 51.92, | |
| "learning_rate": 5.108163265306123e-06, | |
| "loss": 0.8165, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 53.84, | |
| "learning_rate": 4.904081632653061e-06, | |
| "loss": 0.8045, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 55.77, | |
| "learning_rate": 4.7e-06, | |
| "loss": 0.7974, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "learning_rate": 4.4959183673469394e-06, | |
| "loss": 0.7794, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 57.69, | |
| "eval_loss": 0.5800846815109253, | |
| "eval_runtime": 63.4747, | |
| "eval_samples_per_second": 17.519, | |
| "eval_steps_per_second": 4.38, | |
| "eval_wer": 0.7388340807174888, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 59.61, | |
| "learning_rate": 4.291836734693878e-06, | |
| "loss": 0.7752, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 61.54, | |
| "learning_rate": 4.087755102040817e-06, | |
| "loss": 0.7765, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 63.46, | |
| "learning_rate": 3.8836734693877556e-06, | |
| "loss": 0.7584, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 65.38, | |
| "learning_rate": 3.6795918367346943e-06, | |
| "loss": 0.7589, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 67.31, | |
| "learning_rate": 3.475510204081633e-06, | |
| "loss": 0.7565, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 67.31, | |
| "eval_loss": 0.582656979560852, | |
| "eval_runtime": 71.1025, | |
| "eval_samples_per_second": 15.639, | |
| "eval_steps_per_second": 3.91, | |
| "eval_wer": 0.7183856502242153, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 69.23, | |
| "learning_rate": 3.2714285714285717e-06, | |
| "loss": 0.7595, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 71.15, | |
| "learning_rate": 3.0673469387755104e-06, | |
| "loss": 0.747, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 73.08, | |
| "learning_rate": 2.863265306122449e-06, | |
| "loss": 0.7373, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 2.6591836734693882e-06, | |
| "loss": 0.7268, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "learning_rate": 2.455102040816327e-06, | |
| "loss": 0.7303, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "eval_loss": 0.580017626285553, | |
| "eval_runtime": 63.495, | |
| "eval_samples_per_second": 17.513, | |
| "eval_steps_per_second": 4.378, | |
| "eval_wer": 0.7296860986547086, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 78.84, | |
| "learning_rate": 2.2510204081632656e-06, | |
| "loss": 0.7239, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 80.77, | |
| "learning_rate": 2.0469387755102044e-06, | |
| "loss": 0.7201, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 82.69, | |
| "learning_rate": 1.8428571428571428e-06, | |
| "loss": 0.7185, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 84.61, | |
| "learning_rate": 1.6387755102040818e-06, | |
| "loss": 0.7149, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 86.54, | |
| "learning_rate": 1.4346938775510205e-06, | |
| "loss": 0.7128, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 86.54, | |
| "eval_loss": 0.5977014899253845, | |
| "eval_runtime": 64.1516, | |
| "eval_samples_per_second": 17.334, | |
| "eval_steps_per_second": 4.333, | |
| "eval_wer": 0.7149775784753363, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 88.46, | |
| "learning_rate": 1.2306122448979594e-06, | |
| "loss": 0.705, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 90.38, | |
| "learning_rate": 1.026530612244898e-06, | |
| "loss": 0.7129, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 92.31, | |
| "learning_rate": 8.224489795918368e-07, | |
| "loss": 0.7076, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 94.23, | |
| "learning_rate": 6.183673469387755e-07, | |
| "loss": 0.7126, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 96.15, | |
| "learning_rate": 4.142857142857143e-07, | |
| "loss": 0.6972, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 96.15, | |
| "eval_loss": 0.5900700092315674, | |
| "eval_runtime": 63.2743, | |
| "eval_samples_per_second": 17.574, | |
| "eval_steps_per_second": 4.394, | |
| "eval_wer": 0.7176681614349776, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 98.08, | |
| "learning_rate": 2.1020408163265306e-07, | |
| "loss": 0.7001, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 6.1224489795918365e-09, | |
| "loss": 0.6965, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 5200, | |
| "total_flos": 1.4653197993726655e+20, | |
| "train_loss": 1.2273276989276594, | |
| "train_runtime": 32753.7935, | |
| "train_samples_per_second": 10.185, | |
| "train_steps_per_second": 0.159 | |
| } | |
| ], | |
| "max_steps": 5200, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.4653197993726655e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |