| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 35370, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.929318631608708e-05, | |
| "loss": 0.5887, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.858637263217416e-05, | |
| "loss": 0.6212, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.787955894826124e-05, | |
| "loss": 0.6428, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.717274526434832e-05, | |
| "loss": 0.6492, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.64659315804354e-05, | |
| "loss": 0.643, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.575911789652248e-05, | |
| "loss": 0.6446, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.505230421260956e-05, | |
| "loss": 0.6578, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 4.434549052869664e-05, | |
| "loss": 0.6545, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.3638676844783716e-05, | |
| "loss": 0.6706, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.2931863160870795e-05, | |
| "loss": 0.673, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.222504947695788e-05, | |
| "loss": 0.6592, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 4.151823579304496e-05, | |
| "loss": 0.6771, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.081142210913204e-05, | |
| "loss": 0.6889, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.0104608425219116e-05, | |
| "loss": 0.6697, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 30.7024, | |
| "eval_chrf++": 57.2458, | |
| "eval_gen_len": 27.1368, | |
| "eval_loss": 1.2473633289337158, | |
| "eval_runtime": 1116.0922, | |
| "eval_samples_per_second": 4.387, | |
| "eval_spbleu": 42.8416, | |
| "eval_steps_per_second": 0.627, | |
| "eval_ter": 59.9898, | |
| "step": 7074 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 3.9397794741306194e-05, | |
| "loss": 0.4754, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.869098105739327e-05, | |
| "loss": 0.4397, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.798416737348035e-05, | |
| "loss": 0.4587, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.727735368956743e-05, | |
| "loss": 0.4694, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3.6570540005654515e-05, | |
| "loss": 0.458, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 3.5863726321741594e-05, | |
| "loss": 0.4628, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.515691263782867e-05, | |
| "loss": 0.4549, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 3.445009895391575e-05, | |
| "loss": 0.4579, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.374328527000283e-05, | |
| "loss": 0.4775, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 3.303647158608991e-05, | |
| "loss": 0.4674, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.2329657902176986e-05, | |
| "loss": 0.4659, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.162284421826407e-05, | |
| "loss": 0.4693, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 3.091603053435115e-05, | |
| "loss": 0.4554, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 3.0209216850438226e-05, | |
| "loss": 0.4671, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 31.3008, | |
| "eval_chrf++": 57.3245, | |
| "eval_gen_len": 26.9416, | |
| "eval_loss": 1.328170895576477, | |
| "eval_runtime": 1121.14, | |
| "eval_samples_per_second": 4.367, | |
| "eval_spbleu": 43.1798, | |
| "eval_steps_per_second": 0.624, | |
| "eval_ter": 59.0032, | |
| "step": 14148 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 2.9502403166525304e-05, | |
| "loss": 0.3415, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.8795589482612383e-05, | |
| "loss": 0.3063, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.808877579869946e-05, | |
| "loss": 0.3058, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 2.738196211478654e-05, | |
| "loss": 0.3089, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 2.6675148430873625e-05, | |
| "loss": 0.3205, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.5968334746960704e-05, | |
| "loss": 0.3082, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 2.5261521063047782e-05, | |
| "loss": 0.3425, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 2.455470737913486e-05, | |
| "loss": 0.3189, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 2.384789369522194e-05, | |
| "loss": 0.3257, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.3141080011309018e-05, | |
| "loss": 0.3249, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.24342663273961e-05, | |
| "loss": 0.3271, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 2.1727452643483178e-05, | |
| "loss": 0.3362, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 2.1020638959570257e-05, | |
| "loss": 0.3334, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 2.031382527565734e-05, | |
| "loss": 0.3229, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 30.103, | |
| "eval_chrf++": 56.7595, | |
| "eval_gen_len": 27.2512, | |
| "eval_loss": 1.4371376037597656, | |
| "eval_runtime": 1144.2163, | |
| "eval_samples_per_second": 4.279, | |
| "eval_spbleu": 42.008, | |
| "eval_steps_per_second": 0.612, | |
| "eval_ter": 60.4163, | |
| "step": 21222 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 1.9607011591744417e-05, | |
| "loss": 0.2754, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 1.8900197907831496e-05, | |
| "loss": 0.2294, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 1.8193384223918574e-05, | |
| "loss": 0.2256, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 1.7486570540005656e-05, | |
| "loss": 0.2248, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 1.6779756856092735e-05, | |
| "loss": 0.2268, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 1.6072943172179813e-05, | |
| "loss": 0.2215, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1.5366129488266895e-05, | |
| "loss": 0.2233, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 1.4659315804353974e-05, | |
| "loss": 0.2305, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 1.3952502120441052e-05, | |
| "loss": 0.2212, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 1.3245688436528131e-05, | |
| "loss": 0.2248, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 1.2538874752615213e-05, | |
| "loss": 0.2161, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 1.1832061068702292e-05, | |
| "loss": 0.2367, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.112524738478937e-05, | |
| "loss": 0.2326, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.041843370087645e-05, | |
| "loss": 0.2257, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 29.9341, | |
| "eval_chrf++": 56.6787, | |
| "eval_gen_len": 27.2226, | |
| "eval_loss": 1.529828429222107, | |
| "eval_runtime": 1139.5818, | |
| "eval_samples_per_second": 4.296, | |
| "eval_spbleu": 42.0167, | |
| "eval_steps_per_second": 0.614, | |
| "eval_ter": 60.6963, | |
| "step": 28296 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 9.711620016963529e-06, | |
| "loss": 0.2022, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 9.004806333050609e-06, | |
| "loss": 0.1638, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 8.297992649137688e-06, | |
| "loss": 0.1778, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 7.591178965224767e-06, | |
| "loss": 0.1752, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 6.884365281311847e-06, | |
| "loss": 0.1678, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 6.177551597398926e-06, | |
| "loss": 0.1723, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 5.470737913486006e-06, | |
| "loss": 0.1655, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 4.763924229573085e-06, | |
| "loss": 0.1658, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 4.057110545660164e-06, | |
| "loss": 0.1776, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 3.350296861747244e-06, | |
| "loss": 0.1632, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.643483177834323e-06, | |
| "loss": 0.1653, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 1.9366694939214022e-06, | |
| "loss": 0.1639, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 1.2298558100084818e-06, | |
| "loss": 0.1655, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 5.230421260955612e-07, | |
| "loss": 0.1722, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 30.6094, | |
| "eval_chrf++": 56.9952, | |
| "eval_gen_len": 27.0588, | |
| "eval_loss": 1.5824342966079712, | |
| "eval_runtime": 1129.1112, | |
| "eval_samples_per_second": 4.336, | |
| "eval_spbleu": 42.6439, | |
| "eval_steps_per_second": 0.62, | |
| "eval_ter": 59.923, | |
| "step": 35370 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 35370, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2.107220927048581e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |