| { | |
| "best_metric": 0.7927327277273106, | |
| "best_model_checkpoint": "./XLM-V_96-multi-outputs/checkpoint-9000", | |
| "epoch": 14.8619957537155, | |
| "eval_steps": 1000, | |
| "global_step": 14000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0615711252653928, | |
| "grad_norm": 3.787013530731201, | |
| "learning_rate": 1.4154281670205236e-06, | |
| "loss": 0.6909, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0615711252653928, | |
| "eval_accuracy": 0.6001194862093, | |
| "eval_f1": 0.7023421286688408, | |
| "eval_loss": 0.659694254398346, | |
| "eval_precision": 0.5653937947494033, | |
| "eval_recall": 0.9268388106416275, | |
| "eval_runtime": 59.6001, | |
| "eval_samples_per_second": 168.506, | |
| "eval_steps_per_second": 1.762, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.1231422505307855, | |
| "grad_norm": 10.855875015258789, | |
| "learning_rate": 1.907682629550995e-06, | |
| "loss": 0.6595, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.1231422505307855, | |
| "eval_accuracy": 0.6375584984566365, | |
| "eval_f1": 0.727830118139674, | |
| "eval_loss": 0.6210178136825562, | |
| "eval_precision": 0.5890825465988865, | |
| "eval_recall": 0.952073552425665, | |
| "eval_runtime": 59.4085, | |
| "eval_samples_per_second": 169.05, | |
| "eval_steps_per_second": 1.767, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.1847133757961785, | |
| "grad_norm": 1.80464768409729, | |
| "learning_rate": 1.7504128332153809e-06, | |
| "loss": 0.6382, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.1847133757961785, | |
| "eval_accuracy": 0.6707159215373892, | |
| "eval_f1": 0.7504339295147536, | |
| "eval_loss": 0.5838897228240967, | |
| "eval_precision": 0.6108858582135398, | |
| "eval_recall": 0.9726134585289515, | |
| "eval_runtime": 59.4982, | |
| "eval_samples_per_second": 168.795, | |
| "eval_steps_per_second": 1.765, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.246284501061571, | |
| "grad_norm": 3.50860595703125, | |
| "learning_rate": 1.593143036879767e-06, | |
| "loss": 0.6155, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.246284501061571, | |
| "eval_accuracy": 0.6774868067310564, | |
| "eval_f1": 0.752426813421998, | |
| "eval_loss": 0.5712079405784607, | |
| "eval_precision": 0.6174883954334462, | |
| "eval_recall": 0.9628325508607198, | |
| "eval_runtime": 59.4454, | |
| "eval_samples_per_second": 168.945, | |
| "eval_steps_per_second": 1.766, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.307855626326964, | |
| "grad_norm": 10.334639549255371, | |
| "learning_rate": 1.4358732405441535e-06, | |
| "loss": 0.5986, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.307855626326964, | |
| "eval_accuracy": 0.7027780543662252, | |
| "eval_f1": 0.7685867121482285, | |
| "eval_loss": 0.5451286435127258, | |
| "eval_precision": 0.6365737768074997, | |
| "eval_recall": 0.969679186228482, | |
| "eval_runtime": 59.5608, | |
| "eval_samples_per_second": 168.618, | |
| "eval_steps_per_second": 1.763, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.369426751592357, | |
| "grad_norm": 4.514008045196533, | |
| "learning_rate": 1.2786034442085396e-06, | |
| "loss": 0.5815, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.369426751592357, | |
| "eval_accuracy": 0.7189086926217266, | |
| "eval_f1": 0.7648479800083299, | |
| "eval_loss": 0.527675986289978, | |
| "eval_precision": 0.6660380095749311, | |
| "eval_recall": 0.8980829420970265, | |
| "eval_runtime": 59.5312, | |
| "eval_samples_per_second": 168.702, | |
| "eval_steps_per_second": 1.764, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.43099787685775, | |
| "grad_norm": 15.843857765197754, | |
| "learning_rate": 1.121333647872926e-06, | |
| "loss": 0.5596, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 7.43099787685775, | |
| "eval_accuracy": 0.7367320521756447, | |
| "eval_f1": 0.7817401353805514, | |
| "eval_loss": 0.5013874769210815, | |
| "eval_precision": 0.6762353613253356, | |
| "eval_recall": 0.9262519561815337, | |
| "eval_runtime": 59.9272, | |
| "eval_samples_per_second": 167.587, | |
| "eval_steps_per_second": 1.752, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.492569002123142, | |
| "grad_norm": 26.940380096435547, | |
| "learning_rate": 9.640638515373122e-07, | |
| "loss": 0.5332, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 8.492569002123142, | |
| "eval_accuracy": 0.7425072189584785, | |
| "eval_f1": 0.7898244473342002, | |
| "eval_loss": 0.4773547947406769, | |
| "eval_precision": 0.6756117908787542, | |
| "eval_recall": 0.9505086071987481, | |
| "eval_runtime": 59.4845, | |
| "eval_samples_per_second": 168.834, | |
| "eval_steps_per_second": 1.765, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.554140127388536, | |
| "grad_norm": 15.124056816101074, | |
| "learning_rate": 8.067940552016985e-07, | |
| "loss": 0.4836, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.554140127388536, | |
| "eval_accuracy": 0.7523648312257294, | |
| "eval_f1": 0.7927327277273106, | |
| "eval_loss": 0.4266820549964905, | |
| "eval_precision": 0.6905764483810077, | |
| "eval_recall": 0.9303599374021909, | |
| "eval_runtime": 59.8735, | |
| "eval_samples_per_second": 167.737, | |
| "eval_steps_per_second": 1.754, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.615711252653927, | |
| "grad_norm": 11.718687057495117, | |
| "learning_rate": 6.495242588660847e-07, | |
| "loss": 0.4479, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 10.615711252653927, | |
| "eval_accuracy": 0.7561485611868963, | |
| "eval_f1": 0.7909874541264829, | |
| "eval_loss": 0.40309131145477295, | |
| "eval_precision": 0.7015897047691143, | |
| "eval_recall": 0.9064945226917058, | |
| "eval_runtime": 59.4764, | |
| "eval_samples_per_second": 168.857, | |
| "eval_steps_per_second": 1.765, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 11.67728237791932, | |
| "grad_norm": 10.142501831054688, | |
| "learning_rate": 4.92254462530471e-07, | |
| "loss": 0.4328, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 11.67728237791932, | |
| "eval_accuracy": 0.7591357164193966, | |
| "eval_f1": 0.7869285651369682, | |
| "eval_loss": 0.3995474874973297, | |
| "eval_precision": 0.7157506809806121, | |
| "eval_recall": 0.8738262910798122, | |
| "eval_runtime": 59.6214, | |
| "eval_samples_per_second": 168.446, | |
| "eval_steps_per_second": 1.761, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 12.738853503184714, | |
| "grad_norm": 5.893862247467041, | |
| "learning_rate": 3.3498466619485727e-07, | |
| "loss": 0.4217, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.738853503184714, | |
| "eval_accuracy": 0.7629194463805635, | |
| "eval_f1": 0.783683110747706, | |
| "eval_loss": 0.3929121494293213, | |
| "eval_precision": 0.7316369804919424, | |
| "eval_recall": 0.8437010954616588, | |
| "eval_runtime": 59.4335, | |
| "eval_samples_per_second": 168.979, | |
| "eval_steps_per_second": 1.767, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 13.800424628450106, | |
| "grad_norm": 18.15483283996582, | |
| "learning_rate": 1.7771486985924354e-07, | |
| "loss": 0.4159, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 13.800424628450106, | |
| "eval_accuracy": 0.7652095987254804, | |
| "eval_f1": 0.7918799646954987, | |
| "eval_loss": 0.39117059111595154, | |
| "eval_precision": 0.7214538436796397, | |
| "eval_recall": 0.8775430359937402, | |
| "eval_runtime": 59.605, | |
| "eval_samples_per_second": 168.493, | |
| "eval_steps_per_second": 1.762, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 14.8619957537155, | |
| "grad_norm": 9.832246780395508, | |
| "learning_rate": 2.0445073523629785e-08, | |
| "loss": 0.4121, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 14.8619957537155, | |
| "eval_accuracy": 0.7667031763417306, | |
| "eval_f1": 0.7908222480135703, | |
| "eval_loss": 0.3898768126964569, | |
| "eval_precision": 0.7273772376416489, | |
| "eval_recall": 0.8663928012519562, | |
| "eval_runtime": 59.4519, | |
| "eval_samples_per_second": 168.926, | |
| "eval_steps_per_second": 1.766, | |
| "step": 14000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 14130, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.2803191582601536e+17, | |
| "train_batch_size": 96, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |