| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.9873817034700316, | |
| "eval_steps": 500, | |
| "global_step": 632, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 7.812499999999999e-08, | |
| "loss": 4.9285, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.5624999999999999e-07, | |
| "loss": 3.6302, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.3437499999999998e-07, | |
| "loss": 1.2999, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "loss": 0.4051, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 3.9062499999999997e-07, | |
| "loss": 0.1864, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4.6874999999999996e-07, | |
| "loss": 0.1536, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.998623501539504e-07, | |
| "loss": 0.1154, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.990217055187362e-07, | |
| "loss": 0.0672, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.974194562818019e-07, | |
| "loss": 0.0571, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.950605027404507e-07, | |
| "loss": 0.0619, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.91952059486152e-07, | |
| "loss": 0.0554, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.881036333395328e-07, | |
| "loss": 0.0549, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.835269942748475e-07, | |
| "loss": 0.0542, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.782361394228472e-07, | |
| "loss": 0.0527, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.7224725026214615e-07, | |
| "loss": 0.0559, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.0518389493227005, | |
| "eval_runtime": 251.2611, | |
| "eval_samples_per_second": 0.82, | |
| "eval_steps_per_second": 0.072, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 4.655786431300069e-07, | |
| "loss": 0.0447, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.582507132039034e-07, | |
| "loss": 0.0459, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.5028587212518697e-07, | |
| "loss": 0.0442, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.4170847945562717e-07, | |
| "loss": 0.0491, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 4.325447681764586e-07, | |
| "loss": 0.0446, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.228227644577875e-07, | |
| "loss": 0.0449, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.1257220194373424e-07, | |
| "loss": 0.0395, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.0182443081545917e-07, | |
| "loss": 0.0423, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 3.9061232191019517e-07, | |
| "loss": 0.0487, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 3.78970166189525e-07, | |
| "loss": 0.0396, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 3.669335698643704e-07, | |
| "loss": 0.039, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 3.5453934549744043e-07, | |
| "loss": 0.0516, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.418253994161892e-07, | |
| "loss": 0.0431, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 3.288306157806193e-07, | |
| "loss": 0.0405, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3.1559473766049476e-07, | |
| "loss": 0.039, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.021582454856766e-07, | |
| "loss": 0.0426, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.04264240339398384, | |
| "eval_runtime": 248.0203, | |
| "eval_samples_per_second": 0.831, | |
| "eval_steps_per_second": 0.073, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 2.8856223324132555e-07, | |
| "loss": 0.0438, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.748482827866165e-07, | |
| "loss": 0.0355, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.610583366813447e-07, | |
| "loss": 0.0296, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 2.472345699093711e-07, | |
| "loss": 0.0308, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 2.3341926089122408e-07, | |
| "loss": 0.0305, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 2.1965466218035267e-07, | |
| "loss": 0.0357, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.0598287123849092e-07, | |
| "loss": 0.0308, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.9244570168535297e-07, | |
| "loss": 0.0371, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 1.7908455541642582e-07, | |
| "loss": 0.0333, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 1.659402959799753e-07, | |
| "loss": 0.0287, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 1.5305312360052443e-07, | |
| "loss": 0.0316, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 1.4046245223103348e-07, | |
| "loss": 0.0321, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.2820678900980092e-07, | |
| "loss": 0.0281, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.1632361649075498e-07, | |
| "loss": 0.0301, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.0484927800731982e-07, | |
| "loss": 0.028, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 9.381886652045845e-08, | |
| "loss": 0.0405, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.04665624722838402, | |
| "eval_runtime": 248.6648, | |
| "eval_samples_per_second": 0.828, | |
| "eval_steps_per_second": 0.072, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.32661172908373e-08, | |
| "loss": 0.027, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 7.322330470336313e-08, | |
| "loss": 0.0229, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 6.372114355964292e-08, | |
| "loss": 0.0229, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 5.4788695140251637e-08, | |
| "loss": 0.0222, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 4.645327832410648e-08, | |
| "loss": 0.021, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 3.874038603677882e-08, | |
| "loss": 0.0225, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 3.167360728327681e-08, | |
| "loss": 0.0204, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.5274555003752697e-08, | |
| "loss": 0.0206, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 1.956279997278043e-08, | |
| "loss": 0.0225, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 1.4555810944364478e-08, | |
| "loss": 0.0289, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 1.0268901225739979e-08, | |
| "loss": 0.0254, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 6.715181843361617e-09, | |
| "loss": 0.0217, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 3.905521444318604e-09, | |
| "loss": 0.025, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 1.8485130558120454e-09, | |
| "loss": 0.0179, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 5.504478043572291e-10, | |
| "loss": 0.0226, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 1.5295675087678705e-11, | |
| "loss": 0.0158, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_loss": 0.050866369158029556, | |
| "eval_runtime": 243.3946, | |
| "eval_samples_per_second": 0.846, | |
| "eval_steps_per_second": 0.074, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "step": 632, | |
| "total_flos": 72638842798080.0, | |
| "train_loss": 0.20211912293128576, | |
| "train_runtime": 41419.4173, | |
| "train_samples_per_second": 0.367, | |
| "train_steps_per_second": 0.015 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 632, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 72638842798080.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |