| { | |
| "best_metric": 0.7707859873771667, | |
| "best_model_checkpoint": "bert-balanced-v1/checkpoint-204", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 204, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04411764705882353, | |
| "grad_norm": 11.873502731323242, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 1.1054, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08823529411764706, | |
| "grad_norm": 6.837568283081055, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 1.1623, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1323529411764706, | |
| "grad_norm": 6.079131126403809, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 1.1727, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.17647058823529413, | |
| "grad_norm": 7.644276142120361, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.1962, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.22058823529411764, | |
| "grad_norm": 6.449653625488281, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.1334, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2647058823529412, | |
| "grad_norm": 3.3152713775634766, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 1.0897, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3088235294117647, | |
| "grad_norm": 6.6442036628723145, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1438, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.35294117647058826, | |
| "grad_norm": 3.3698086738586426, | |
| "learning_rate": 4.918032786885246e-05, | |
| "loss": 1.2006, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.39705882352941174, | |
| "grad_norm": 7.311341762542725, | |
| "learning_rate": 4.836065573770492e-05, | |
| "loss": 1.0572, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": 8.757562637329102, | |
| "learning_rate": 4.754098360655738e-05, | |
| "loss": 1.0659, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4852941176470588, | |
| "grad_norm": 7.169471263885498, | |
| "learning_rate": 4.672131147540984e-05, | |
| "loss": 1.1263, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5294117647058824, | |
| "grad_norm": 6.60593843460083, | |
| "learning_rate": 4.59016393442623e-05, | |
| "loss": 1.0724, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5735294117647058, | |
| "grad_norm": 5.335102558135986, | |
| "learning_rate": 4.508196721311476e-05, | |
| "loss": 1.0535, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.6176470588235294, | |
| "grad_norm": 5.638067245483398, | |
| "learning_rate": 4.426229508196721e-05, | |
| "loss": 1.0671, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6617647058823529, | |
| "grad_norm": 3.3761277198791504, | |
| "learning_rate": 4.3442622950819674e-05, | |
| "loss": 1.0608, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 3.7545318603515625, | |
| "learning_rate": 4.262295081967213e-05, | |
| "loss": 1.036, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 3.6131749153137207, | |
| "learning_rate": 4.1803278688524595e-05, | |
| "loss": 1.0085, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7941176470588235, | |
| "grad_norm": 10.428824424743652, | |
| "learning_rate": 4.098360655737705e-05, | |
| "loss": 0.9663, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8382352941176471, | |
| "grad_norm": 9.748165130615234, | |
| "learning_rate": 4.016393442622951e-05, | |
| "loss": 0.9838, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 7.34658145904541, | |
| "learning_rate": 3.934426229508197e-05, | |
| "loss": 0.914, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9264705882352942, | |
| "grad_norm": 9.839737892150879, | |
| "learning_rate": 3.8524590163934424e-05, | |
| "loss": 1.0774, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.9705882352941176, | |
| "grad_norm": 6.513272762298584, | |
| "learning_rate": 3.7704918032786885e-05, | |
| "loss": 0.9173, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6074074074074074, | |
| "eval_f1_macro": 0.5499705029992206, | |
| "eval_f1_micro": 0.6074074074074074, | |
| "eval_f1_weighted": 0.5499705029992207, | |
| "eval_loss": 0.9292003512382507, | |
| "eval_precision_macro": 0.6655672369958084, | |
| "eval_precision_micro": 0.6074074074074074, | |
| "eval_precision_weighted": 0.6655672369958084, | |
| "eval_recall_macro": 0.6074074074074074, | |
| "eval_recall_micro": 0.6074074074074074, | |
| "eval_recall_weighted": 0.6074074074074074, | |
| "eval_runtime": 134.9252, | |
| "eval_samples_per_second": 1.001, | |
| "eval_steps_per_second": 0.067, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.0147058823529411, | |
| "grad_norm": 6.397573947906494, | |
| "learning_rate": 3.6885245901639346e-05, | |
| "loss": 1.1385, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.0588235294117647, | |
| "grad_norm": 8.004409790039062, | |
| "learning_rate": 3.6065573770491806e-05, | |
| "loss": 0.8346, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.1029411764705883, | |
| "grad_norm": 9.197546005249023, | |
| "learning_rate": 3.524590163934427e-05, | |
| "loss": 0.7762, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.1470588235294117, | |
| "grad_norm": 15.531254768371582, | |
| "learning_rate": 3.442622950819672e-05, | |
| "loss": 1.0938, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.1911764705882353, | |
| "grad_norm": 9.45050048828125, | |
| "learning_rate": 3.360655737704918e-05, | |
| "loss": 0.7187, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.2352941176470589, | |
| "grad_norm": 7.438510417938232, | |
| "learning_rate": 3.2786885245901635e-05, | |
| "loss": 0.6956, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.2794117647058822, | |
| "grad_norm": 7.671905994415283, | |
| "learning_rate": 3.19672131147541e-05, | |
| "loss": 0.8591, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": 5.862545013427734, | |
| "learning_rate": 3.114754098360656e-05, | |
| "loss": 0.7878, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.3676470588235294, | |
| "grad_norm": 12.08448314666748, | |
| "learning_rate": 3.0327868852459017e-05, | |
| "loss": 0.922, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 25.875642776489258, | |
| "learning_rate": 2.9508196721311478e-05, | |
| "loss": 0.806, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.4558823529411764, | |
| "grad_norm": 7.914124488830566, | |
| "learning_rate": 2.8688524590163935e-05, | |
| "loss": 0.7999, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 7.366003513336182, | |
| "learning_rate": 2.7868852459016392e-05, | |
| "loss": 0.8753, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.5441176470588234, | |
| "grad_norm": 7.287258625030518, | |
| "learning_rate": 2.7049180327868856e-05, | |
| "loss": 0.75, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.5882352941176472, | |
| "grad_norm": 6.466761112213135, | |
| "learning_rate": 2.6229508196721314e-05, | |
| "loss": 0.7331, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.6323529411764706, | |
| "grad_norm": 11.213692665100098, | |
| "learning_rate": 2.540983606557377e-05, | |
| "loss": 0.9052, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.6764705882352942, | |
| "grad_norm": 10.924762725830078, | |
| "learning_rate": 2.459016393442623e-05, | |
| "loss": 1.0455, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.7205882352941178, | |
| "grad_norm": 8.067938804626465, | |
| "learning_rate": 2.377049180327869e-05, | |
| "loss": 0.8233, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 6.702548027038574, | |
| "learning_rate": 2.295081967213115e-05, | |
| "loss": 1.0777, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.8088235294117647, | |
| "grad_norm": 5.71842098236084, | |
| "learning_rate": 2.2131147540983607e-05, | |
| "loss": 0.6789, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 1.8529411764705883, | |
| "grad_norm": 8.926857948303223, | |
| "learning_rate": 2.1311475409836064e-05, | |
| "loss": 0.6347, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.8970588235294117, | |
| "grad_norm": 21.031246185302734, | |
| "learning_rate": 2.0491803278688525e-05, | |
| "loss": 0.7921, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 1.9411764705882353, | |
| "grad_norm": 12.221565246582031, | |
| "learning_rate": 1.9672131147540985e-05, | |
| "loss": 0.8242, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.9852941176470589, | |
| "grad_norm": 4.548130035400391, | |
| "learning_rate": 1.8852459016393442e-05, | |
| "loss": 0.5987, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6888888888888889, | |
| "eval_f1_macro": 0.6845081103855066, | |
| "eval_f1_micro": 0.6888888888888889, | |
| "eval_f1_weighted": 0.6845081103855067, | |
| "eval_loss": 0.811522901058197, | |
| "eval_precision_macro": 0.7045918367346938, | |
| "eval_precision_micro": 0.6888888888888889, | |
| "eval_precision_weighted": 0.7045918367346938, | |
| "eval_recall_macro": 0.6888888888888888, | |
| "eval_recall_micro": 0.6888888888888889, | |
| "eval_recall_weighted": 0.6888888888888889, | |
| "eval_runtime": 131.5615, | |
| "eval_samples_per_second": 1.026, | |
| "eval_steps_per_second": 0.068, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.0294117647058822, | |
| "grad_norm": 7.7553510665893555, | |
| "learning_rate": 1.8032786885245903e-05, | |
| "loss": 0.6751, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 2.073529411764706, | |
| "grad_norm": 3.660639762878418, | |
| "learning_rate": 1.721311475409836e-05, | |
| "loss": 0.637, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 7.876177787780762, | |
| "learning_rate": 1.6393442622950818e-05, | |
| "loss": 0.6454, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.161764705882353, | |
| "grad_norm": 8.173316955566406, | |
| "learning_rate": 1.557377049180328e-05, | |
| "loss": 0.5053, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 7.380105018615723, | |
| "learning_rate": 1.4754098360655739e-05, | |
| "loss": 0.64, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 13.381223678588867, | |
| "learning_rate": 1.3934426229508196e-05, | |
| "loss": 0.7163, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.2941176470588234, | |
| "grad_norm": 21.8228759765625, | |
| "learning_rate": 1.3114754098360657e-05, | |
| "loss": 0.8269, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.338235294117647, | |
| "grad_norm": 6.742076396942139, | |
| "learning_rate": 1.2295081967213116e-05, | |
| "loss": 0.5206, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 2.3823529411764706, | |
| "grad_norm": 8.277242660522461, | |
| "learning_rate": 1.1475409836065575e-05, | |
| "loss": 0.5042, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 2.426470588235294, | |
| "grad_norm": 9.727288246154785, | |
| "learning_rate": 1.0655737704918032e-05, | |
| "loss": 0.6658, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.4705882352941178, | |
| "grad_norm": 18.33995819091797, | |
| "learning_rate": 9.836065573770493e-06, | |
| "loss": 0.6377, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 2.514705882352941, | |
| "grad_norm": 8.533559799194336, | |
| "learning_rate": 9.016393442622952e-06, | |
| "loss": 0.8016, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 2.5588235294117645, | |
| "grad_norm": 22.230371475219727, | |
| "learning_rate": 8.196721311475409e-06, | |
| "loss": 0.6268, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 2.6029411764705883, | |
| "grad_norm": 7.892202854156494, | |
| "learning_rate": 7.3770491803278695e-06, | |
| "loss": 0.732, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 2.6470588235294117, | |
| "grad_norm": 6.714171886444092, | |
| "learning_rate": 6.557377049180328e-06, | |
| "loss": 0.6977, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.6911764705882355, | |
| "grad_norm": 5.960449695587158, | |
| "learning_rate": 5.737704918032787e-06, | |
| "loss": 0.3479, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 2.735294117647059, | |
| "grad_norm": 5.625789165496826, | |
| "learning_rate": 4.918032786885246e-06, | |
| "loss": 0.5536, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.7794117647058822, | |
| "grad_norm": 3.9824419021606445, | |
| "learning_rate": 4.098360655737704e-06, | |
| "loss": 0.6288, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 4.617522716522217, | |
| "learning_rate": 3.278688524590164e-06, | |
| "loss": 0.4954, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.8676470588235294, | |
| "grad_norm": 5.714521408081055, | |
| "learning_rate": 2.459016393442623e-06, | |
| "loss": 0.4388, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.911764705882353, | |
| "grad_norm": 11.925094604492188, | |
| "learning_rate": 1.639344262295082e-06, | |
| "loss": 0.871, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 2.9558823529411766, | |
| "grad_norm": 9.523457527160645, | |
| "learning_rate": 8.19672131147541e-07, | |
| "loss": 0.5172, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 19.425655364990234, | |
| "learning_rate": 0.0, | |
| "loss": 0.4496, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.674074074074074, | |
| "eval_f1_macro": 0.6722359201219389, | |
| "eval_f1_micro": 0.674074074074074, | |
| "eval_f1_weighted": 0.6722359201219389, | |
| "eval_loss": 0.7707859873771667, | |
| "eval_precision_macro": 0.6750194250194249, | |
| "eval_precision_micro": 0.674074074074074, | |
| "eval_precision_weighted": 0.675019425019425, | |
| "eval_recall_macro": 0.674074074074074, | |
| "eval_recall_micro": 0.674074074074074, | |
| "eval_recall_weighted": 0.674074074074074, | |
| "eval_runtime": 139.3011, | |
| "eval_samples_per_second": 0.969, | |
| "eval_steps_per_second": 0.065, | |
| "step": 204 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 204, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 423875715959808.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |