| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 23430, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.893299189073837e-05, | |
| "loss": 0.7701, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.786598378147674e-05, | |
| "loss": 0.1282, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.6798975672215114e-05, | |
| "loss": 0.0914, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.5731967562953484e-05, | |
| "loss": 0.0791, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.4664959453691854e-05, | |
| "loss": 0.0656, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.359795134443022e-05, | |
| "loss": 0.0645, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.253094323516859e-05, | |
| "loss": 0.0573, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.146393512590696e-05, | |
| "loss": 0.0566, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.039692701664533e-05, | |
| "loss": 0.0544, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 25.989, | |
| "eval_gen_len": 18.8655, | |
| "eval_loss": 0.044239919632673264, | |
| "eval_runtime": 180.9818, | |
| "eval_samples_per_second": 5.752, | |
| "eval_steps_per_second": 2.879, | |
| "step": 4686 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.93299189073837e-05, | |
| "loss": 0.0476, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 3.826291079812207e-05, | |
| "loss": 0.0442, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.719590268886043e-05, | |
| "loss": 0.0375, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 3.6128894579598804e-05, | |
| "loss": 0.0408, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 3.5061886470337174e-05, | |
| "loss": 0.0351, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.3994878361075544e-05, | |
| "loss": 0.0355, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 3.2927870251813915e-05, | |
| "loss": 0.0379, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 3.1860862142552285e-05, | |
| "loss": 0.0331, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 3.0793854033290656e-05, | |
| "loss": 0.0367, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 25.9358, | |
| "eval_gen_len": 18.8713, | |
| "eval_loss": 0.03710315003991127, | |
| "eval_runtime": 179.552, | |
| "eval_samples_per_second": 5.798, | |
| "eval_steps_per_second": 2.902, | |
| "step": 9372 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 2.9726845924029023e-05, | |
| "loss": 0.0307, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 2.8659837814767393e-05, | |
| "loss": 0.0218, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.7592829705505763e-05, | |
| "loss": 0.0234, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 2.6525821596244134e-05, | |
| "loss": 0.0265, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 2.54588134869825e-05, | |
| "loss": 0.0239, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2.439180537772087e-05, | |
| "loss": 0.0261, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.332479726845924e-05, | |
| "loss": 0.0247, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 2.2257789159197612e-05, | |
| "loss": 0.0228, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 2.1190781049935982e-05, | |
| "loss": 0.0233, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 2.0123772940674353e-05, | |
| "loss": 0.0222, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 25.8976, | |
| "eval_gen_len": 18.8694, | |
| "eval_loss": 0.037382081151008606, | |
| "eval_runtime": 178.2936, | |
| "eval_samples_per_second": 5.839, | |
| "eval_steps_per_second": 2.922, | |
| "step": 14058 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 1.905676483141272e-05, | |
| "loss": 0.0156, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.7989756722151087e-05, | |
| "loss": 0.0153, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 1.6922748612889457e-05, | |
| "loss": 0.0174, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 1.5855740503627827e-05, | |
| "loss": 0.0171, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 1.4788732394366198e-05, | |
| "loss": 0.0176, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 1.3721724285104568e-05, | |
| "loss": 0.0187, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 1.2654716175842937e-05, | |
| "loss": 0.0157, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.1587708066581307e-05, | |
| "loss": 0.0169, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 1.0520699957319676e-05, | |
| "loss": 0.0152, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 26.1575, | |
| "eval_gen_len": 18.8694, | |
| "eval_loss": 0.040931396186351776, | |
| "eval_runtime": 182.3446, | |
| "eval_samples_per_second": 5.709, | |
| "eval_steps_per_second": 2.857, | |
| "step": 18744 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 9.453691848058044e-06, | |
| "loss": 0.0148, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 8.386683738796415e-06, | |
| "loss": 0.0124, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 7.319675629534785e-06, | |
| "loss": 0.0148, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 6.252667520273155e-06, | |
| "loss": 0.0123, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 5.1856594110115235e-06, | |
| "loss": 0.0139, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 4.118651301749894e-06, | |
| "loss": 0.0116, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 3.051643192488263e-06, | |
| "loss": 0.014, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 1.9846350832266325e-06, | |
| "loss": 0.0143, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 9.176269739650021e-07, | |
| "loss": 0.0147, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 26.0973, | |
| "eval_gen_len": 18.8694, | |
| "eval_loss": 0.043235890567302704, | |
| "eval_runtime": 184.0328, | |
| "eval_samples_per_second": 5.657, | |
| "eval_steps_per_second": 2.831, | |
| "step": 23430 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 23430, | |
| "total_flos": 2019762272885760.0, | |
| "train_loss": 0.047856709064706696, | |
| "train_runtime": 5577.5722, | |
| "train_samples_per_second": 8.401, | |
| "train_steps_per_second": 4.201 | |
| } | |
| ], | |
| "max_steps": 23430, | |
| "num_train_epochs": 5, | |
| "total_flos": 2019762272885760.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |