| { | |
| "best_metric": 0.03381425514817238, | |
| "best_model_checkpoint": "autotrain-pmf0g-rj8fa/checkpoint-330", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 330, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.5416135787963867, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 1.3908, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.7443188428878784, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.3651, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 2.275134563446045, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 1.2519, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 2.0600032806396484, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.9604, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 1.9959027767181396, | |
| "learning_rate": 4.454545454545455e-05, | |
| "loss": 0.5255, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9748858447488584, | |
| "eval_f1_macro": 0.9756765342178342, | |
| "eval_f1_micro": 0.9748858447488584, | |
| "eval_f1_weighted": 0.9749751427585824, | |
| "eval_loss": 0.20833216607570648, | |
| "eval_precision_macro": 0.9759725400457666, | |
| "eval_precision_micro": 0.9748858447488584, | |
| "eval_precision_weighted": 0.9751052736068879, | |
| "eval_recall_macro": 0.9754215648989755, | |
| "eval_recall_micro": 0.9748858447488584, | |
| "eval_recall_weighted": 0.9748858447488584, | |
| "eval_runtime": 0.1296, | |
| "eval_samples_per_second": 3380.791, | |
| "eval_steps_per_second": 54.031, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 2.120863199234009, | |
| "learning_rate": 4.9595959595959594e-05, | |
| "loss": 0.2706, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 0.6585132479667664, | |
| "learning_rate": 4.858585858585859e-05, | |
| "loss": 0.1525, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 4.553682327270508, | |
| "learning_rate": 4.7575757575757576e-05, | |
| "loss": 0.0592, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.41076862812042236, | |
| "learning_rate": 4.656565656565657e-05, | |
| "loss": 0.0615, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 3.347062110900879, | |
| "learning_rate": 4.565656565656566e-05, | |
| "loss": 0.1097, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.071051597595215, | |
| "learning_rate": 4.464646464646465e-05, | |
| "loss": 0.0851, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9748858447488584, | |
| "eval_f1_macro": 0.9754834367482361, | |
| "eval_f1_micro": 0.9748858447488584, | |
| "eval_f1_weighted": 0.9749041744765222, | |
| "eval_loss": 0.07598946988582611, | |
| "eval_precision_macro": 0.9755606462503015, | |
| "eval_precision_micro": 0.9748858447488584, | |
| "eval_precision_weighted": 0.9752209332157372, | |
| "eval_recall_macro": 0.9756989697593826, | |
| "eval_recall_micro": 0.9748858447488584, | |
| "eval_recall_weighted": 0.9748858447488584, | |
| "eval_runtime": 0.1241, | |
| "eval_samples_per_second": 3530.172, | |
| "eval_steps_per_second": 56.418, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 0.41413745284080505, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.0477, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 5.531299591064453, | |
| "learning_rate": 4.262626262626263e-05, | |
| "loss": 0.0332, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 0.4154397249221802, | |
| "learning_rate": 4.161616161616162e-05, | |
| "loss": 0.0398, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 2.7882883548736572, | |
| "learning_rate": 4.0606060606060606e-05, | |
| "loss": 0.027, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 3.329517364501953, | |
| "learning_rate": 3.9595959595959594e-05, | |
| "loss": 0.0104, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9885844748858448, | |
| "eval_f1_macro": 0.9888365654549831, | |
| "eval_f1_micro": 0.9885844748858448, | |
| "eval_f1_weighted": 0.9886138949055431, | |
| "eval_loss": 0.04520614817738533, | |
| "eval_precision_macro": 0.9889521258788082, | |
| "eval_precision_micro": 0.9885844748858448, | |
| "eval_precision_weighted": 0.9888428242322514, | |
| "eval_recall_macro": 0.9889150852693966, | |
| "eval_recall_micro": 0.9885844748858448, | |
| "eval_recall_weighted": 0.9885844748858448, | |
| "eval_runtime": 0.1261, | |
| "eval_samples_per_second": 3473.478, | |
| "eval_steps_per_second": 55.512, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 0.08116048574447632, | |
| "learning_rate": 3.858585858585859e-05, | |
| "loss": 0.0334, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 0.1431998461484909, | |
| "learning_rate": 3.757575757575758e-05, | |
| "loss": 0.0244, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "grad_norm": 0.04078197479248047, | |
| "learning_rate": 3.656565656565657e-05, | |
| "loss": 0.0069, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 0.03086886927485466, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.0042, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "grad_norm": 0.03221021592617035, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 0.0324, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.04350810497999191, | |
| "learning_rate": 3.3535353535353536e-05, | |
| "loss": 0.0203, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9840182648401826, | |
| "eval_f1_macro": 0.9844112150421562, | |
| "eval_f1_micro": 0.9840182648401826, | |
| "eval_f1_weighted": 0.9839758921897677, | |
| "eval_loss": 0.06017115339636803, | |
| "eval_precision_macro": 0.9844066713333537, | |
| "eval_precision_micro": 0.9840182648401826, | |
| "eval_precision_weighted": 0.984073353425936, | |
| "eval_recall_macro": 0.9845520881850645, | |
| "eval_recall_micro": 0.9840182648401826, | |
| "eval_recall_weighted": 0.9840182648401826, | |
| "eval_runtime": 0.124, | |
| "eval_samples_per_second": 3533.459, | |
| "eval_steps_per_second": 56.471, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "grad_norm": 0.09443140029907227, | |
| "learning_rate": 3.2525252525252524e-05, | |
| "loss": 0.0034, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "grad_norm": 2.7363362312316895, | |
| "learning_rate": 3.151515151515151e-05, | |
| "loss": 0.0049, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 0.026695426553487778, | |
| "learning_rate": 3.050505050505051e-05, | |
| "loss": 0.0024, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "grad_norm": 0.023897232487797737, | |
| "learning_rate": 2.9494949494949498e-05, | |
| "loss": 0.0022, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "grad_norm": 0.021171841770410538, | |
| "learning_rate": 2.8484848484848486e-05, | |
| "loss": 0.0021, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9885844748858448, | |
| "eval_f1_macro": 0.9888140343800068, | |
| "eval_f1_micro": 0.9885844748858448, | |
| "eval_f1_weighted": 0.9885396362946075, | |
| "eval_loss": 0.06403101235628128, | |
| "eval_precision_macro": 0.9890982392627942, | |
| "eval_precision_micro": 0.9885844748858448, | |
| "eval_precision_weighted": 0.9889162145164453, | |
| "eval_recall_macro": 0.9889380530973452, | |
| "eval_recall_micro": 0.9885844748858448, | |
| "eval_recall_weighted": 0.9885844748858448, | |
| "eval_runtime": 0.1253, | |
| "eval_samples_per_second": 3496.411, | |
| "eval_steps_per_second": 55.879, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 0.018654897809028625, | |
| "learning_rate": 2.7474747474747474e-05, | |
| "loss": 0.0024, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "grad_norm": 0.02241847664117813, | |
| "learning_rate": 2.6464646464646466e-05, | |
| "loss": 0.0019, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "grad_norm": 0.018668798729777336, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.0017, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "grad_norm": 0.020943326875567436, | |
| "learning_rate": 2.4444444444444445e-05, | |
| "loss": 0.0017, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "grad_norm": 0.01833685301244259, | |
| "learning_rate": 2.3434343434343436e-05, | |
| "loss": 0.0017, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.01634540781378746, | |
| "learning_rate": 2.2424242424242424e-05, | |
| "loss": 0.0016, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9908675799086758, | |
| "eval_f1_macro": 0.9910410929202866, | |
| "eval_f1_micro": 0.9908675799086758, | |
| "eval_f1_weighted": 0.9908473335613555, | |
| "eval_loss": 0.03381425514817238, | |
| "eval_precision_macro": 0.9909727371947719, | |
| "eval_precision_micro": 0.9908675799086758, | |
| "eval_precision_weighted": 0.9908883151237302, | |
| "eval_recall_macro": 0.9911698494022667, | |
| "eval_recall_micro": 0.9908675799086758, | |
| "eval_recall_weighted": 0.9908675799086758, | |
| "eval_runtime": 0.124, | |
| "eval_samples_per_second": 3531.36, | |
| "eval_steps_per_second": 56.437, | |
| "step": 330 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 349726407917568.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |