| { | |
| "best_metric": 1.2287747859954834, | |
| "best_model_checkpoint": "./outputs/checkpoint-3900", | |
| "epoch": 2.841894353369763, | |
| "eval_steps": 100, | |
| "global_step": 3900, | |
| "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": 2.1823, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.0118842124938965, | |
| "eval_runtime": 144.2983, | |
| "eval_samples_per_second": 43.479, | |
| "eval_steps_per_second": 5.44, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 1.962, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.9193025827407837, | |
| "eval_runtime": 144.1022, | |
| "eval_samples_per_second": 43.539, | |
| "eval_steps_per_second": 5.448, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8883, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.8596361875534058, | |
| "eval_runtime": 144.0831, | |
| "eval_samples_per_second": 43.544, | |
| "eval_steps_per_second": 5.448, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8371, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.813263177871704, | |
| "eval_runtime": 144.1028, | |
| "eval_samples_per_second": 43.538, | |
| "eval_steps_per_second": 5.447, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7855, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.772437572479248, | |
| "eval_runtime": 144.0494, | |
| "eval_samples_per_second": 43.555, | |
| "eval_steps_per_second": 5.45, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.757, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.7428079843521118, | |
| "eval_runtime": 144.0319, | |
| "eval_samples_per_second": 43.56, | |
| "eval_steps_per_second": 5.45, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7183, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.7120596170425415, | |
| "eval_runtime": 144.1455, | |
| "eval_samples_per_second": 43.525, | |
| "eval_steps_per_second": 5.446, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6973, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.6833879947662354, | |
| "eval_runtime": 144.034, | |
| "eval_samples_per_second": 43.559, | |
| "eval_steps_per_second": 5.45, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.662, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.6580077409744263, | |
| "eval_runtime": 144.0204, | |
| "eval_samples_per_second": 43.563, | |
| "eval_steps_per_second": 5.451, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6473, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.6349676847457886, | |
| "eval_runtime": 144.1987, | |
| "eval_samples_per_second": 43.509, | |
| "eval_steps_per_second": 5.444, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6273, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.6135053634643555, | |
| "eval_runtime": 144.1005, | |
| "eval_samples_per_second": 43.539, | |
| "eval_steps_per_second": 5.448, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5919, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.5944637060165405, | |
| "eval_runtime": 144.0899, | |
| "eval_samples_per_second": 43.542, | |
| "eval_steps_per_second": 5.448, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5994, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.5728504657745361, | |
| "eval_runtime": 144.043, | |
| "eval_samples_per_second": 43.556, | |
| "eval_steps_per_second": 5.45, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5528, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.552846074104309, | |
| "eval_runtime": 144.0891, | |
| "eval_samples_per_second": 43.543, | |
| "eval_steps_per_second": 5.448, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5246, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.5355615615844727, | |
| "eval_runtime": 144.0408, | |
| "eval_samples_per_second": 43.557, | |
| "eval_steps_per_second": 5.45, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5062, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.5179370641708374, | |
| "eval_runtime": 144.0242, | |
| "eval_samples_per_second": 43.562, | |
| "eval_steps_per_second": 5.45, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5038, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.5012134313583374, | |
| "eval_runtime": 144.0166, | |
| "eval_samples_per_second": 43.564, | |
| "eval_steps_per_second": 5.451, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5144, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.511275291442871, | |
| "eval_runtime": 133.6708, | |
| "eval_samples_per_second": 46.936, | |
| "eval_steps_per_second": 5.873, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4715, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.48880934715271, | |
| "eval_runtime": 133.6301, | |
| "eval_samples_per_second": 46.95, | |
| "eval_steps_per_second": 5.874, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4621, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.4694921970367432, | |
| "eval_runtime": 133.7845, | |
| "eval_samples_per_second": 46.896, | |
| "eval_steps_per_second": 5.868, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4364, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.4534579515457153, | |
| "eval_runtime": 133.5069, | |
| "eval_samples_per_second": 46.994, | |
| "eval_steps_per_second": 5.88, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4388, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.4382610321044922, | |
| "eval_runtime": 133.6146, | |
| "eval_samples_per_second": 46.956, | |
| "eval_steps_per_second": 5.875, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4139, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.4250658750534058, | |
| "eval_runtime": 133.6118, | |
| "eval_samples_per_second": 46.957, | |
| "eval_steps_per_second": 5.875, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4145, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.408768892288208, | |
| "eval_runtime": 133.5488, | |
| "eval_samples_per_second": 46.979, | |
| "eval_steps_per_second": 5.878, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3897, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 1.3950382471084595, | |
| "eval_runtime": 133.6055, | |
| "eval_samples_per_second": 46.959, | |
| "eval_steps_per_second": 5.876, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3718, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.38124418258667, | |
| "eval_runtime": 133.4827, | |
| "eval_samples_per_second": 47.002, | |
| "eval_steps_per_second": 5.881, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3685, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.3680918216705322, | |
| "eval_runtime": 133.3719, | |
| "eval_samples_per_second": 47.041, | |
| "eval_steps_per_second": 5.886, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3321, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.356438159942627, | |
| "eval_runtime": 133.4175, | |
| "eval_samples_per_second": 47.025, | |
| "eval_steps_per_second": 5.884, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3105, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.3440583944320679, | |
| "eval_runtime": 133.4445, | |
| "eval_samples_per_second": 47.016, | |
| "eval_steps_per_second": 5.883, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3096, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.3319122791290283, | |
| "eval_runtime": 133.4813, | |
| "eval_samples_per_second": 47.003, | |
| "eval_steps_per_second": 5.881, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2963, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 1.320979356765747, | |
| "eval_runtime": 133.3932, | |
| "eval_samples_per_second": 47.034, | |
| "eval_steps_per_second": 5.885, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2953, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.3095366954803467, | |
| "eval_runtime": 138.3583, | |
| "eval_samples_per_second": 45.346, | |
| "eval_steps_per_second": 5.674, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2786, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.2959859371185303, | |
| "eval_runtime": 133.5245, | |
| "eval_samples_per_second": 46.988, | |
| "eval_steps_per_second": 5.879, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2585, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.2855974435806274, | |
| "eval_runtime": 133.542, | |
| "eval_samples_per_second": 46.981, | |
| "eval_steps_per_second": 5.878, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2586, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 1.2743042707443237, | |
| "eval_runtime": 133.3944, | |
| "eval_samples_per_second": 47.033, | |
| "eval_steps_per_second": 5.885, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2466, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 1.2626945972442627, | |
| "eval_runtime": 133.3673, | |
| "eval_samples_per_second": 47.043, | |
| "eval_steps_per_second": 5.886, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2303, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.2512463331222534, | |
| "eval_runtime": 133.4502, | |
| "eval_samples_per_second": 47.014, | |
| "eval_steps_per_second": 5.882, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2088, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.2429386377334595, | |
| "eval_runtime": 133.4443, | |
| "eval_samples_per_second": 47.016, | |
| "eval_steps_per_second": 5.883, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2171, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.2287747859954834, | |
| "eval_runtime": 133.4716, | |
| "eval_samples_per_second": 47.006, | |
| "eval_steps_per_second": 5.881, | |
| "step": 3900 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.3246378200834867e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |