| { | |
| "best_global_step": 425, | |
| "best_metric": 0.9176470588235294, | |
| "best_model_checkpoint": "./bert-samudra-model\\checkpoint-425", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 425, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 10.0350341796875, | |
| "learning_rate": 1.9894117647058826e-05, | |
| "loss": 3.8456, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 10.431258201599121, | |
| "learning_rate": 1.9776470588235296e-05, | |
| "loss": 3.826, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 11.249832153320312, | |
| "learning_rate": 1.9658823529411765e-05, | |
| "loss": 3.6472, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 10.304410934448242, | |
| "learning_rate": 1.9541176470588235e-05, | |
| "loss": 3.5359, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 11.847091674804688, | |
| "learning_rate": 1.942352941176471e-05, | |
| "loss": 3.3157, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 9.621623039245605, | |
| "learning_rate": 1.930588235294118e-05, | |
| "loss": 3.1498, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8235294117647058, | |
| "grad_norm": 11.151117324829102, | |
| "learning_rate": 1.918823529411765e-05, | |
| "loss": 2.9821, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 9.83305549621582, | |
| "learning_rate": 1.907058823529412e-05, | |
| "loss": 2.9091, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5294117647058824, | |
| "eval_loss": 2.518275737762451, | |
| "eval_runtime": 2.8699, | |
| "eval_samples_per_second": 59.235, | |
| "eval_steps_per_second": 7.666, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 9.598007202148438, | |
| "learning_rate": 1.895294117647059e-05, | |
| "loss": 2.5816, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 9.898895263671875, | |
| "learning_rate": 1.883529411764706e-05, | |
| "loss": 2.3487, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2941176470588236, | |
| "grad_norm": 11.571386337280273, | |
| "learning_rate": 1.871764705882353e-05, | |
| "loss": 2.1136, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 11.180049896240234, | |
| "learning_rate": 1.86e-05, | |
| "loss": 2.0731, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5294117647058822, | |
| "grad_norm": 9.434500694274902, | |
| "learning_rate": 1.848235294117647e-05, | |
| "loss": 1.9983, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 8.993762969970703, | |
| "learning_rate": 1.836470588235294e-05, | |
| "loss": 1.7445, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 9.424057960510254, | |
| "learning_rate": 1.8247058823529415e-05, | |
| "loss": 1.7651, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 8.946263313293457, | |
| "learning_rate": 1.8129411764705885e-05, | |
| "loss": 1.6938, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 9.5614013671875, | |
| "learning_rate": 1.8011764705882354e-05, | |
| "loss": 1.539, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8117647058823529, | |
| "eval_loss": 1.4311776161193848, | |
| "eval_runtime": 2.9195, | |
| "eval_samples_per_second": 58.23, | |
| "eval_steps_per_second": 7.536, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 8.067319869995117, | |
| "learning_rate": 1.7894117647058824e-05, | |
| "loss": 1.2424, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 9.209125518798828, | |
| "learning_rate": 1.7776470588235294e-05, | |
| "loss": 1.1468, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 9.35853099822998, | |
| "learning_rate": 1.7658823529411768e-05, | |
| "loss": 1.0893, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 7.949193477630615, | |
| "learning_rate": 1.7541176470588238e-05, | |
| "loss": 1.1078, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 10.602166175842285, | |
| "learning_rate": 1.7423529411764707e-05, | |
| "loss": 0.9902, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.7058823529411766, | |
| "grad_norm": 8.58541488647461, | |
| "learning_rate": 1.7305882352941177e-05, | |
| "loss": 0.9298, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 6.725298881530762, | |
| "learning_rate": 1.7188235294117647e-05, | |
| "loss": 0.8838, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 8.761615753173828, | |
| "learning_rate": 1.7070588235294117e-05, | |
| "loss": 0.8112, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.888235294117647, | |
| "eval_loss": 0.8797827959060669, | |
| "eval_runtime": 2.1675, | |
| "eval_samples_per_second": 78.433, | |
| "eval_steps_per_second": 10.15, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 5.102208137512207, | |
| "learning_rate": 1.695294117647059e-05, | |
| "loss": 0.7609, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.176470588235294, | |
| "grad_norm": 8.779169082641602, | |
| "learning_rate": 1.683529411764706e-05, | |
| "loss": 0.6983, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 4.434069633483887, | |
| "learning_rate": 1.671764705882353e-05, | |
| "loss": 0.5764, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.411764705882353, | |
| "grad_norm": 12.731746673583984, | |
| "learning_rate": 1.66e-05, | |
| "loss": 0.5199, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 6.222368240356445, | |
| "learning_rate": 1.648235294117647e-05, | |
| "loss": 0.5333, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.6470588235294117, | |
| "grad_norm": 5.6505560874938965, | |
| "learning_rate": 1.6364705882352944e-05, | |
| "loss": 0.4524, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 5.23166036605835, | |
| "learning_rate": 1.6247058823529413e-05, | |
| "loss": 0.4974, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.8823529411764706, | |
| "grad_norm": 6.096327304840088, | |
| "learning_rate": 1.6129411764705883e-05, | |
| "loss": 0.4621, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 7.395042419433594, | |
| "learning_rate": 1.6011764705882353e-05, | |
| "loss": 0.4012, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8941176470588236, | |
| "eval_loss": 0.5824028253555298, | |
| "eval_runtime": 2.3688, | |
| "eval_samples_per_second": 71.766, | |
| "eval_steps_per_second": 9.287, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 5.869823455810547, | |
| "learning_rate": 1.5894117647058823e-05, | |
| "loss": 0.3211, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 4.795381546020508, | |
| "learning_rate": 1.5776470588235297e-05, | |
| "loss": 0.3647, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.352941176470588, | |
| "grad_norm": 3.4636142253875732, | |
| "learning_rate": 1.5658823529411766e-05, | |
| "loss": 0.2564, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 10.156576156616211, | |
| "learning_rate": 1.5541176470588236e-05, | |
| "loss": 0.2745, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.588235294117647, | |
| "grad_norm": 2.828077554702759, | |
| "learning_rate": 1.5423529411764706e-05, | |
| "loss": 0.2418, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 3.717984199523926, | |
| "learning_rate": 1.5305882352941176e-05, | |
| "loss": 0.2605, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.823529411764706, | |
| "grad_norm": 1.5987720489501953, | |
| "learning_rate": 1.5188235294117648e-05, | |
| "loss": 0.2369, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 3.3276009559631348, | |
| "learning_rate": 1.507058823529412e-05, | |
| "loss": 0.2366, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9176470588235294, | |
| "eval_loss": 0.4427616000175476, | |
| "eval_runtime": 3.6872, | |
| "eval_samples_per_second": 46.106, | |
| "eval_steps_per_second": 5.967, | |
| "step": 425 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "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": 29670808006140.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |