| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.471204188481675, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.25e-06, | |
| "loss": 7.3722, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_runtime": 4.9957, | |
| "eval_samples_per_second": 20.017, | |
| "eval_steps_per_second": 10.009, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 9.895833333333334e-06, | |
| "loss": 1.5052, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_runtime": 5.5484, | |
| "eval_samples_per_second": 18.023, | |
| "eval_steps_per_second": 9.012, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 9.635416666666668e-06, | |
| "loss": 1.3479, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_runtime": 5.4316, | |
| "eval_samples_per_second": 18.411, | |
| "eval_steps_per_second": 9.205, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 1.3177, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_runtime": 5.4246, | |
| "eval_samples_per_second": 18.435, | |
| "eval_steps_per_second": 9.217, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 9.114583333333334e-06, | |
| "loss": 1.2962, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_runtime": 5.0933, | |
| "eval_samples_per_second": 19.634, | |
| "eval_steps_per_second": 9.817, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 8.854166666666667e-06, | |
| "loss": 1.2824, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_runtime": 6.1894, | |
| "eval_samples_per_second": 16.157, | |
| "eval_steps_per_second": 8.078, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 8.59375e-06, | |
| "loss": 1.2784, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_runtime": 4.7573, | |
| "eval_samples_per_second": 21.02, | |
| "eval_steps_per_second": 10.51, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.2655, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "eval_runtime": 5.5685, | |
| "eval_samples_per_second": 17.958, | |
| "eval_steps_per_second": 8.979, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.072916666666667e-06, | |
| "loss": 1.2553, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_runtime": 6.732, | |
| "eval_samples_per_second": 14.854, | |
| "eval_steps_per_second": 7.427, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 7.8125e-06, | |
| "loss": 1.2627, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_runtime": 4.5523, | |
| "eval_samples_per_second": 21.967, | |
| "eval_steps_per_second": 10.983, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 7.552083333333334e-06, | |
| "loss": 1.2358, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_runtime": 4.906, | |
| "eval_samples_per_second": 20.383, | |
| "eval_steps_per_second": 10.192, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 7.291666666666667e-06, | |
| "loss": 1.2495, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_runtime": 5.2379, | |
| "eval_samples_per_second": 19.092, | |
| "eval_steps_per_second": 9.546, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 7.031250000000001e-06, | |
| "loss": 1.234, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_runtime": 5.2459, | |
| "eval_samples_per_second": 19.062, | |
| "eval_steps_per_second": 9.531, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 6.770833333333334e-06, | |
| "loss": 1.2228, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "eval_runtime": 5.186, | |
| "eval_samples_per_second": 19.283, | |
| "eval_steps_per_second": 9.641, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 6.510416666666667e-06, | |
| "loss": 1.2403, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "eval_runtime": 5.9443, | |
| "eval_samples_per_second": 16.823, | |
| "eval_steps_per_second": 8.411, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 6.25e-06, | |
| "loss": 1.2254, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_runtime": 6.1004, | |
| "eval_samples_per_second": 16.392, | |
| "eval_steps_per_second": 8.196, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 5.989583333333334e-06, | |
| "loss": 1.2203, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "eval_runtime": 4.7155, | |
| "eval_samples_per_second": 21.207, | |
| "eval_steps_per_second": 10.603, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 5.729166666666667e-06, | |
| "loss": 1.2198, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "eval_runtime": 5.2917, | |
| "eval_samples_per_second": 18.897, | |
| "eval_steps_per_second": 9.449, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 5.468750000000001e-06, | |
| "loss": 1.2405, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "eval_runtime": 5.6681, | |
| "eval_samples_per_second": 17.643, | |
| "eval_steps_per_second": 8.821, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 5.208333333333334e-06, | |
| "loss": 1.2209, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "eval_runtime": 4.7064, | |
| "eval_samples_per_second": 21.247, | |
| "eval_steps_per_second": 10.624, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 4.947916666666667e-06, | |
| "loss": 1.232, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_runtime": 5.1725, | |
| "eval_samples_per_second": 19.333, | |
| "eval_steps_per_second": 9.667, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 4.6875000000000004e-06, | |
| "loss": 1.2078, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "eval_runtime": 4.845, | |
| "eval_samples_per_second": 20.64, | |
| "eval_steps_per_second": 10.32, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 4.427083333333334e-06, | |
| "loss": 1.2066, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "eval_runtime": 5.6538, | |
| "eval_samples_per_second": 17.687, | |
| "eval_steps_per_second": 8.844, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.2265, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "eval_runtime": 5.06, | |
| "eval_samples_per_second": 19.763, | |
| "eval_steps_per_second": 9.881, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 3.90625e-06, | |
| "loss": 1.2113, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "eval_runtime": 5.8625, | |
| "eval_samples_per_second": 17.057, | |
| "eval_steps_per_second": 8.529, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3.6458333333333333e-06, | |
| "loss": 1.1924, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "eval_runtime": 5.4916, | |
| "eval_samples_per_second": 18.21, | |
| "eval_steps_per_second": 9.105, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 3.385416666666667e-06, | |
| "loss": 1.2023, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "eval_runtime": 6.0325, | |
| "eval_samples_per_second": 16.577, | |
| "eval_steps_per_second": 8.288, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 3.125e-06, | |
| "loss": 1.2005, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "eval_runtime": 5.7214, | |
| "eval_samples_per_second": 17.478, | |
| "eval_steps_per_second": 8.739, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 2.8645833333333334e-06, | |
| "loss": 1.2046, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_runtime": 5.7834, | |
| "eval_samples_per_second": 17.291, | |
| "eval_steps_per_second": 8.645, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 2.604166666666667e-06, | |
| "loss": 1.2132, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "eval_runtime": 6.0375, | |
| "eval_samples_per_second": 16.563, | |
| "eval_steps_per_second": 8.282, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 2.3437500000000002e-06, | |
| "loss": 1.2044, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "eval_runtime": 5.1804, | |
| "eval_samples_per_second": 19.304, | |
| "eval_steps_per_second": 9.652, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 2.0833333333333334e-06, | |
| "loss": 1.1986, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "eval_runtime": 7.2562, | |
| "eval_samples_per_second": 13.781, | |
| "eval_steps_per_second": 6.891, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "learning_rate": 1.8229166666666666e-06, | |
| "loss": 1.2057, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.64, | |
| "eval_runtime": 4.8338, | |
| "eval_samples_per_second": 20.688, | |
| "eval_steps_per_second": 10.344, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 1.5625e-06, | |
| "loss": 1.1947, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_runtime": 5.2971, | |
| "eval_samples_per_second": 18.878, | |
| "eval_steps_per_second": 9.439, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 1.3020833333333335e-06, | |
| "loss": 1.215, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "eval_runtime": 5.1412, | |
| "eval_samples_per_second": 19.451, | |
| "eval_steps_per_second": 9.725, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 1.0416666666666667e-06, | |
| "loss": 1.191, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "eval_runtime": 4.8814, | |
| "eval_samples_per_second": 20.486, | |
| "eval_steps_per_second": 10.243, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "learning_rate": 7.8125e-07, | |
| "loss": 1.2013, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "eval_runtime": 5.6388, | |
| "eval_samples_per_second": 17.734, | |
| "eval_steps_per_second": 8.867, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 5.208333333333334e-07, | |
| "loss": 1.1838, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "eval_runtime": 8.057, | |
| "eval_samples_per_second": 12.412, | |
| "eval_steps_per_second": 6.206, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 2.604166666666667e-07, | |
| "loss": 1.2097, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "eval_runtime": 8.8555, | |
| "eval_samples_per_second": 11.292, | |
| "eval_steps_per_second": 5.646, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 0.0, | |
| "loss": 1.1946, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "eval_runtime": 7.1071, | |
| "eval_samples_per_second": 14.07, | |
| "eval_steps_per_second": 7.035, | |
| "step": 20000 | |
| } | |
| ], | |
| "max_steps": 20000, | |
| "num_train_epochs": 11, | |
| "total_flos": 1.673264670519214e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |