| { | |
| "best_metric": 0.8202670216560364, | |
| "best_model_checkpoint": "not_included_mt5-base/checkpoint-5000", | |
| "epoch": 5.9880239520958085, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11976047904191617, | |
| "grad_norm": 91.10150146484375, | |
| "learning_rate": 0.0001976047904191617, | |
| "loss": 10.8476, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23952095808383234, | |
| "grad_norm": 3.626096248626709, | |
| "learning_rate": 0.00019520958083832338, | |
| "loss": 1.901, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3592814371257485, | |
| "grad_norm": 1.472233772277832, | |
| "learning_rate": 0.00019281437125748504, | |
| "loss": 1.5652, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.47904191616766467, | |
| "grad_norm": 1.3357157707214355, | |
| "learning_rate": 0.0001904191616766467, | |
| "loss": 1.3415, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5988023952095808, | |
| "grad_norm": 0.947651207447052, | |
| "learning_rate": 0.0001880239520958084, | |
| "loss": 1.233, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5988023952095808, | |
| "eval_Accuracy": 0.062, | |
| "eval_loss": 1.1174869537353516, | |
| "eval_runtime": 22.1894, | |
| "eval_samples_per_second": 225.333, | |
| "eval_steps_per_second": 7.075, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.718562874251497, | |
| "grad_norm": 0.8490985035896301, | |
| "learning_rate": 0.00018562874251497007, | |
| "loss": 1.1601, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8383233532934131, | |
| "grad_norm": 0.8438432812690735, | |
| "learning_rate": 0.00018323353293413173, | |
| "loss": 1.1129, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.9580838323353293, | |
| "grad_norm": 0.7390198707580566, | |
| "learning_rate": 0.00018083832335329342, | |
| "loss": 1.0631, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.0778443113772456, | |
| "grad_norm": 0.8146846890449524, | |
| "learning_rate": 0.0001784431137724551, | |
| "loss": 1.0243, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.1976047904191618, | |
| "grad_norm": 0.7685344815254211, | |
| "learning_rate": 0.0001760479041916168, | |
| "loss": 0.9898, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.1976047904191618, | |
| "eval_Accuracy": 0.0998, | |
| "eval_loss": 0.9793121814727783, | |
| "eval_runtime": 22.9009, | |
| "eval_samples_per_second": 218.332, | |
| "eval_steps_per_second": 6.856, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3173652694610778, | |
| "grad_norm": 0.7814993262290955, | |
| "learning_rate": 0.00017365269461077845, | |
| "loss": 0.9735, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.437125748502994, | |
| "grad_norm": 0.7339932322502136, | |
| "learning_rate": 0.0001712574850299401, | |
| "loss": 0.9723, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.55688622754491, | |
| "grad_norm": 0.7049471735954285, | |
| "learning_rate": 0.0001688622754491018, | |
| "loss": 0.9567, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.6766467065868262, | |
| "grad_norm": 0.7287872433662415, | |
| "learning_rate": 0.00016646706586826348, | |
| "loss": 0.946, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.7964071856287425, | |
| "grad_norm": 0.6998888254165649, | |
| "learning_rate": 0.00016407185628742517, | |
| "loss": 0.9204, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.7964071856287425, | |
| "eval_Accuracy": 0.1246, | |
| "eval_loss": 0.9095961451530457, | |
| "eval_runtime": 23.1954, | |
| "eval_samples_per_second": 215.56, | |
| "eval_steps_per_second": 6.769, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9161676646706587, | |
| "grad_norm": 0.9079245328903198, | |
| "learning_rate": 0.00016167664670658683, | |
| "loss": 0.9237, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.035928143712575, | |
| "grad_norm": 0.6102410554885864, | |
| "learning_rate": 0.0001592814371257485, | |
| "loss": 0.9064, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.155688622754491, | |
| "grad_norm": 0.6484932899475098, | |
| "learning_rate": 0.00015688622754491018, | |
| "loss": 0.8548, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.2754491017964074, | |
| "grad_norm": 0.6946207880973816, | |
| "learning_rate": 0.00015449101796407186, | |
| "loss": 0.8632, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.3952095808383236, | |
| "grad_norm": 0.706642210483551, | |
| "learning_rate": 0.00015209580838323355, | |
| "loss": 0.8405, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.3952095808383236, | |
| "eval_Accuracy": 0.1388, | |
| "eval_loss": 0.8839116096496582, | |
| "eval_runtime": 22.9368, | |
| "eval_samples_per_second": 217.99, | |
| "eval_steps_per_second": 6.845, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.5149700598802394, | |
| "grad_norm": 0.7250691056251526, | |
| "learning_rate": 0.0001497005988023952, | |
| "loss": 0.8546, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.6347305389221556, | |
| "grad_norm": 0.7567930817604065, | |
| "learning_rate": 0.0001473053892215569, | |
| "loss": 0.8508, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.754491017964072, | |
| "grad_norm": 0.8484090566635132, | |
| "learning_rate": 0.00014491017964071858, | |
| "loss": 0.851, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.874251497005988, | |
| "grad_norm": 0.8208755254745483, | |
| "learning_rate": 0.00014251497005988024, | |
| "loss": 0.843, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.9940119760479043, | |
| "grad_norm": 0.6463755965232849, | |
| "learning_rate": 0.00014011976047904193, | |
| "loss": 0.8338, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.9940119760479043, | |
| "eval_Accuracy": 0.1382, | |
| "eval_loss": 0.852607786655426, | |
| "eval_runtime": 23.4307, | |
| "eval_samples_per_second": 213.396, | |
| "eval_steps_per_second": 6.701, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.1137724550898205, | |
| "grad_norm": 0.708572506904602, | |
| "learning_rate": 0.00013772455089820359, | |
| "loss": 0.7838, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.2335329341317367, | |
| "grad_norm": 0.758407711982727, | |
| "learning_rate": 0.00013532934131736527, | |
| "loss": 0.7792, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.3532934131736525, | |
| "grad_norm": 0.8018171191215515, | |
| "learning_rate": 0.00013293413173652696, | |
| "loss": 0.7873, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.4730538922155687, | |
| "grad_norm": 0.6300553679466248, | |
| "learning_rate": 0.00013053892215568865, | |
| "loss": 0.774, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.592814371257485, | |
| "grad_norm": 0.5713552832603455, | |
| "learning_rate": 0.0001281437125748503, | |
| "loss": 0.783, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.592814371257485, | |
| "eval_Accuracy": 0.1508, | |
| "eval_loss": 0.8408400416374207, | |
| "eval_runtime": 23.0015, | |
| "eval_samples_per_second": 217.377, | |
| "eval_steps_per_second": 6.826, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.712574850299401, | |
| "grad_norm": 0.6296516060829163, | |
| "learning_rate": 0.00012574850299401196, | |
| "loss": 0.7694, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.8323353293413174, | |
| "grad_norm": 0.5926547050476074, | |
| "learning_rate": 0.00012335329341317365, | |
| "loss": 0.7878, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.9520958083832336, | |
| "grad_norm": 0.8825701475143433, | |
| "learning_rate": 0.00012095808383233534, | |
| "loss": 0.7683, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.07185628742515, | |
| "grad_norm": 0.7714102268218994, | |
| "learning_rate": 0.00011856287425149701, | |
| "loss": 0.7411, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.191616766467066, | |
| "grad_norm": 0.7947181463241577, | |
| "learning_rate": 0.0001161676646706587, | |
| "loss": 0.7288, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.191616766467066, | |
| "eval_Accuracy": 0.1602, | |
| "eval_loss": 0.8330999612808228, | |
| "eval_runtime": 23.0702, | |
| "eval_samples_per_second": 216.73, | |
| "eval_steps_per_second": 6.805, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.311377245508982, | |
| "grad_norm": 0.7584651112556458, | |
| "learning_rate": 0.00011377245508982037, | |
| "loss": 0.7258, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.431137724550898, | |
| "grad_norm": 0.6492884159088135, | |
| "learning_rate": 0.00011137724550898203, | |
| "loss": 0.7222, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.550898203592815, | |
| "grad_norm": 0.7623780965805054, | |
| "learning_rate": 0.00010898203592814371, | |
| "loss": 0.7168, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.6706586826347305, | |
| "grad_norm": 0.8104670643806458, | |
| "learning_rate": 0.00010658682634730539, | |
| "loss": 0.7272, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.790419161676647, | |
| "grad_norm": 0.6871969699859619, | |
| "learning_rate": 0.00010419161676646707, | |
| "loss": 0.7204, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.790419161676647, | |
| "eval_Accuracy": 0.159, | |
| "eval_loss": 0.8263267874717712, | |
| "eval_runtime": 22.9773, | |
| "eval_samples_per_second": 217.606, | |
| "eval_steps_per_second": 6.833, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.910179640718563, | |
| "grad_norm": 0.7578293085098267, | |
| "learning_rate": 0.00010179640718562875, | |
| "loss": 0.727, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.029940119760479, | |
| "grad_norm": 0.71566241979599, | |
| "learning_rate": 9.940119760479042e-05, | |
| "loss": 0.7111, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.149700598802395, | |
| "grad_norm": 0.7744144201278687, | |
| "learning_rate": 9.700598802395209e-05, | |
| "loss": 0.6699, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.269461077844311, | |
| "grad_norm": 0.7886649370193481, | |
| "learning_rate": 9.461077844311378e-05, | |
| "loss": 0.6791, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.389221556886228, | |
| "grad_norm": 0.7046491503715515, | |
| "learning_rate": 9.221556886227547e-05, | |
| "loss": 0.6533, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.389221556886228, | |
| "eval_Accuracy": 0.1644, | |
| "eval_loss": 0.8296283483505249, | |
| "eval_runtime": 23.2437, | |
| "eval_samples_per_second": 215.112, | |
| "eval_steps_per_second": 6.755, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.508982035928144, | |
| "grad_norm": 0.6836217641830444, | |
| "learning_rate": 8.982035928143712e-05, | |
| "loss": 0.6798, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 5.62874251497006, | |
| "grad_norm": 0.8316423296928406, | |
| "learning_rate": 8.742514970059881e-05, | |
| "loss": 0.6813, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 5.748502994011976, | |
| "grad_norm": 0.806846022605896, | |
| "learning_rate": 8.502994011976048e-05, | |
| "loss": 0.6797, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.868263473053892, | |
| "grad_norm": 0.7254876494407654, | |
| "learning_rate": 8.263473053892216e-05, | |
| "loss": 0.6732, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 5.9880239520958085, | |
| "grad_norm": 1.0394260883331299, | |
| "learning_rate": 8.023952095808383e-05, | |
| "loss": 0.6955, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.9880239520958085, | |
| "eval_Accuracy": 0.1616, | |
| "eval_loss": 0.8202670216560364, | |
| "eval_runtime": 23.0559, | |
| "eval_samples_per_second": 216.864, | |
| "eval_steps_per_second": 6.81, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 8350, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.01 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4755250684394496.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |