| { | |
| "best_metric": 1.3103234767913818, | |
| "best_model_checkpoint": "./kozh_mbartLarge_50p_run3_tokenize/checkpoint-7500", | |
| "epoch": 7.980845969672785, | |
| "eval_steps": 7500, | |
| "global_step": 30000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.5913, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.8919, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7345, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.930728733721253e-05, | |
| "loss": 1.6451, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.861457467442505e-05, | |
| "loss": 1.5539, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.792186201163758e-05, | |
| "loss": 1.512, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.72291493488501e-05, | |
| "loss": 1.4671, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 4.6536436686062625e-05, | |
| "loss": 1.417, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.5843724023275146e-05, | |
| "loss": 1.3449, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.515101136048767e-05, | |
| "loss": 1.2543, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 4.4458298697700194e-05, | |
| "loss": 1.1794, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.376558603491272e-05, | |
| "loss": 1.1359, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 4.307287337212524e-05, | |
| "loss": 1.1075, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 4.238016070933777e-05, | |
| "loss": 1.1079, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.168744804655029e-05, | |
| "loss": 1.0808, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 51.0184, | |
| "eval_gen_len": 13.3575, | |
| "eval_loss": 1.3103234767913818, | |
| "eval_runtime": 1625.4898, | |
| "eval_samples_per_second": 18.499, | |
| "eval_steps_per_second": 1.157, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 4.099473538376282e-05, | |
| "loss": 1.053, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 4.0302022720975344e-05, | |
| "loss": 0.9994, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 3.9609310058187865e-05, | |
| "loss": 0.9291, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 3.891659739540039e-05, | |
| "loss": 0.8958, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 3.822388473261291e-05, | |
| "loss": 0.8643, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 3.753117206982544e-05, | |
| "loss": 0.8633, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 3.683845940703796e-05, | |
| "loss": 0.8519, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 3.614574674425049e-05, | |
| "loss": 0.8366, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 3.545303408146301e-05, | |
| "loss": 0.8027, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.4760321418675535e-05, | |
| "loss": 0.7519, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.4067608755888056e-05, | |
| "loss": 0.7062, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 3.337489609310058e-05, | |
| "loss": 0.6797, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 3.2682183430313104e-05, | |
| "loss": 0.6629, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 3.198947076752563e-05, | |
| "loss": 0.6719, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 3.129675810473816e-05, | |
| "loss": 0.6519, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_bleu": 51.6667, | |
| "eval_gen_len": 13.4112, | |
| "eval_loss": 1.3856228590011597, | |
| "eval_runtime": 1615.8993, | |
| "eval_samples_per_second": 18.609, | |
| "eval_steps_per_second": 1.163, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 3.060404544195068e-05, | |
| "loss": 0.6434, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 2.9911332779163203e-05, | |
| "loss": 0.6042, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 2.9218620116375727e-05, | |
| "loss": 0.5549, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 2.852590745358825e-05, | |
| "loss": 0.535, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 2.7833194790800778e-05, | |
| "loss": 0.5144, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 2.7140482128013302e-05, | |
| "loss": 0.518, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 2.6447769465225826e-05, | |
| "loss": 0.5116, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 2.575505680243835e-05, | |
| "loss": 0.5012, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 2.5062344139650874e-05, | |
| "loss": 0.4845, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 2.4369631476863397e-05, | |
| "loss": 0.4486, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 2.367691881407592e-05, | |
| "loss": 0.4172, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 2.2984206151288445e-05, | |
| "loss": 0.4, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 2.229149348850097e-05, | |
| "loss": 0.3908, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 2.1598780825713493e-05, | |
| "loss": 0.4034, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 2.0906068162926017e-05, | |
| "loss": 0.3867, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_bleu": 51.5686, | |
| "eval_gen_len": 13.432, | |
| "eval_loss": 1.5243059396743774, | |
| "eval_runtime": 1619.767, | |
| "eval_samples_per_second": 18.564, | |
| "eval_steps_per_second": 1.161, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 2.021335550013854e-05, | |
| "loss": 0.3857, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 1.9520642837351065e-05, | |
| "loss": 0.3601, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 1.8827930174563592e-05, | |
| "loss": 0.3289, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 1.8135217511776116e-05, | |
| "loss": 0.3148, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 1.744250484898864e-05, | |
| "loss": 0.3033, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 1.6749792186201167e-05, | |
| "loss": 0.3056, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 1.605707952341369e-05, | |
| "loss": 0.3055, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 1.5364366860626215e-05, | |
| "loss": 0.2993, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 1.4671654197838739e-05, | |
| "loss": 0.2928, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 1.3978941535051263e-05, | |
| "loss": 0.2674, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 1.3286228872263787e-05, | |
| "loss": 0.2481, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 1.259351620947631e-05, | |
| "loss": 0.2385, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "learning_rate": 1.1900803546688833e-05, | |
| "loss": 0.2339, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 1.1208090883901359e-05, | |
| "loss": 0.2439, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 1.0515378221113883e-05, | |
| "loss": 0.236, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "eval_bleu": 51.4674, | |
| "eval_gen_len": 13.291, | |
| "eval_loss": 1.6158065795898438, | |
| "eval_runtime": 1612.187, | |
| "eval_samples_per_second": 18.652, | |
| "eval_steps_per_second": 1.166, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "step": 30000, | |
| "total_flos": 4.16088915443712e+18, | |
| "train_loss": 0.7653622683207194, | |
| "train_runtime": 99492.172, | |
| "train_samples_per_second": 24.18, | |
| "train_steps_per_second": 0.378 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 37590, | |
| "num_train_epochs": 10, | |
| "save_steps": 7500, | |
| "total_flos": 4.16088915443712e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |