| { | |
| "best_metric": 1.3161581754684448, | |
| "best_model_checkpoint": "./outputs/checkpoint-4100", | |
| "epoch": 2.9879781420765026, | |
| "eval_steps": 100, | |
| "global_step": 4100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3134, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.1594152450561523, | |
| "eval_runtime": 144.083, | |
| "eval_samples_per_second": 43.544, | |
| "eval_steps_per_second": 5.448, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1084, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.06351900100708, | |
| "eval_runtime": 143.9954, | |
| "eval_samples_per_second": 43.571, | |
| "eval_steps_per_second": 5.452, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0343, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.0027661323547363, | |
| "eval_runtime": 144.0229, | |
| "eval_samples_per_second": 43.563, | |
| "eval_steps_per_second": 5.451, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9806, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.9533618688583374, | |
| "eval_runtime": 143.9496, | |
| "eval_samples_per_second": 43.585, | |
| "eval_steps_per_second": 5.453, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.926, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.9100102186203003, | |
| "eval_runtime": 144.0297, | |
| "eval_samples_per_second": 43.56, | |
| "eval_steps_per_second": 5.45, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8969, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.8807153701782227, | |
| "eval_runtime": 144.1421, | |
| "eval_samples_per_second": 43.526, | |
| "eval_steps_per_second": 5.446, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8568, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.8459755182266235, | |
| "eval_runtime": 144.0267, | |
| "eval_samples_per_second": 43.561, | |
| "eval_steps_per_second": 5.45, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8329, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.8178023099899292, | |
| "eval_runtime": 143.9854, | |
| "eval_samples_per_second": 43.574, | |
| "eval_steps_per_second": 5.452, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7961, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.7928833961486816, | |
| "eval_runtime": 144.015, | |
| "eval_samples_per_second": 43.565, | |
| "eval_steps_per_second": 5.451, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7798, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.7671397924423218, | |
| "eval_runtime": 144.0009, | |
| "eval_samples_per_second": 43.569, | |
| "eval_steps_per_second": 5.451, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7596, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.7436497211456299, | |
| "eval_runtime": 144.11, | |
| "eval_samples_per_second": 43.536, | |
| "eval_steps_per_second": 5.447, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7215, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.7219586372375488, | |
| "eval_runtime": 144.0047, | |
| "eval_samples_per_second": 43.568, | |
| "eval_steps_per_second": 5.451, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7295, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.700698733329773, | |
| "eval_runtime": 143.9971, | |
| "eval_samples_per_second": 43.57, | |
| "eval_steps_per_second": 5.451, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6805, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.679167628288269, | |
| "eval_runtime": 144.0847, | |
| "eval_samples_per_second": 43.544, | |
| "eval_steps_per_second": 5.448, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6505, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.6613633632659912, | |
| "eval_runtime": 144.175, | |
| "eval_samples_per_second": 43.517, | |
| "eval_steps_per_second": 5.445, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6327, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.6426126956939697, | |
| "eval_runtime": 143.9637, | |
| "eval_samples_per_second": 43.58, | |
| "eval_steps_per_second": 5.453, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.629, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.6252305507659912, | |
| "eval_runtime": 144.0148, | |
| "eval_samples_per_second": 43.565, | |
| "eval_steps_per_second": 5.451, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5993, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.6091866493225098, | |
| "eval_runtime": 144.0349, | |
| "eval_samples_per_second": 43.559, | |
| "eval_steps_per_second": 5.45, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5907, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.5939242839813232, | |
| "eval_runtime": 151.8105, | |
| "eval_samples_per_second": 41.328, | |
| "eval_steps_per_second": 5.171, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5726, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.578723669052124, | |
| "eval_runtime": 144.0665, | |
| "eval_samples_per_second": 43.549, | |
| "eval_steps_per_second": 5.449, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5572, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.562243938446045, | |
| "eval_runtime": 144.0023, | |
| "eval_samples_per_second": 43.569, | |
| "eval_steps_per_second": 5.451, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5788, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.5802664756774902, | |
| "eval_runtime": 133.2244, | |
| "eval_samples_per_second": 47.093, | |
| "eval_steps_per_second": 5.892, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5406, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.5586791038513184, | |
| "eval_runtime": 133.3887, | |
| "eval_samples_per_second": 47.035, | |
| "eval_steps_per_second": 5.885, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5397, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.5387272834777832, | |
| "eval_runtime": 133.2598, | |
| "eval_samples_per_second": 47.081, | |
| "eval_steps_per_second": 5.891, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5072, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 1.5237348079681396, | |
| "eval_runtime": 133.2857, | |
| "eval_samples_per_second": 47.072, | |
| "eval_steps_per_second": 5.89, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4925, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.5068706274032593, | |
| "eval_runtime": 133.2428, | |
| "eval_samples_per_second": 47.087, | |
| "eval_steps_per_second": 5.891, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4832, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.492112636566162, | |
| "eval_runtime": 133.2152, | |
| "eval_samples_per_second": 47.097, | |
| "eval_steps_per_second": 5.893, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4506, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.4801561832427979, | |
| "eval_runtime": 133.2908, | |
| "eval_samples_per_second": 47.07, | |
| "eval_steps_per_second": 5.889, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4311, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.4649614095687866, | |
| "eval_runtime": 133.232, | |
| "eval_samples_per_second": 47.091, | |
| "eval_steps_per_second": 5.892, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4268, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.4508098363876343, | |
| "eval_runtime": 133.3471, | |
| "eval_samples_per_second": 47.05, | |
| "eval_steps_per_second": 5.887, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4142, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 1.4384832382202148, | |
| "eval_runtime": 133.1612, | |
| "eval_samples_per_second": 47.116, | |
| "eval_steps_per_second": 5.895, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4131, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.4258862733840942, | |
| "eval_runtime": 133.2271, | |
| "eval_samples_per_second": 47.093, | |
| "eval_steps_per_second": 5.892, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3948, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.4105219841003418, | |
| "eval_runtime": 133.2146, | |
| "eval_samples_per_second": 47.097, | |
| "eval_steps_per_second": 5.893, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3738, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.3999289274215698, | |
| "eval_runtime": 133.2454, | |
| "eval_samples_per_second": 47.086, | |
| "eval_steps_per_second": 5.891, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 1.373, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 1.388107180595398, | |
| "eval_runtime": 133.2079, | |
| "eval_samples_per_second": 47.099, | |
| "eval_steps_per_second": 5.893, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3325, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 1.3749202489852905, | |
| "eval_runtime": 132.999, | |
| "eval_samples_per_second": 47.173, | |
| "eval_steps_per_second": 5.902, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.332, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.3657797574996948, | |
| "eval_runtime": 133.0705, | |
| "eval_samples_per_second": 47.148, | |
| "eval_steps_per_second": 5.899, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3103, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.3516861200332642, | |
| "eval_runtime": 132.9998, | |
| "eval_samples_per_second": 47.173, | |
| "eval_steps_per_second": 5.902, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3085, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.340726613998413, | |
| "eval_runtime": 132.9904, | |
| "eval_samples_per_second": 47.176, | |
| "eval_steps_per_second": 5.903, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2963, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 1.3258968591690063, | |
| "eval_runtime": 132.9668, | |
| "eval_samples_per_second": 47.185, | |
| "eval_steps_per_second": 5.904, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2867, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 1.3161581754684448, | |
| "eval_runtime": 132.905, | |
| "eval_samples_per_second": 47.207, | |
| "eval_steps_per_second": 5.906, | |
| "step": 4100 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.4411263908542874e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |