| { | |
| "best_metric": 0.6307053941908713, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-qtr0udgj/checkpoint-200", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.049558639526367, | |
| "learning_rate": 8.243172962606281e-06, | |
| "loss": 0.663, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.8758213520050049, | |
| "learning_rate": 7.809321754048056e-06, | |
| "loss": 0.5872, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.6676493883132935, | |
| "learning_rate": 7.375470545489831e-06, | |
| "loss": 0.5806, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.160999298095703, | |
| "learning_rate": 6.941619336931606e-06, | |
| "loss": 0.5375, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 3.5581603050231934, | |
| "learning_rate": 6.507768128373381e-06, | |
| "loss": 0.5219, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.43050479888916, | |
| "learning_rate": 6.0739169198151545e-06, | |
| "loss": 0.4933, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 3.4380669593811035, | |
| "learning_rate": 5.64006571125693e-06, | |
| "loss": 0.4417, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.310192823410034, | |
| "learning_rate": 5.2062145026987045e-06, | |
| "loss": 0.484, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 5.5063323974609375, | |
| "learning_rate": 4.772363294140479e-06, | |
| "loss": 0.5157, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.589657783508301, | |
| "learning_rate": 4.3385120855822536e-06, | |
| "loss": 0.4096, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.755, | |
| "eval_f1": 0.5882352941176471, | |
| "eval_loss": 0.4866395890712738, | |
| "eval_precision": 0.6086956521739131, | |
| "eval_recall": 0.5691056910569106, | |
| "eval_runtime": 1.5211, | |
| "eval_samples_per_second": 262.961, | |
| "eval_steps_per_second": 16.435, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.2699267864227295, | |
| "learning_rate": 3.904660877024028e-06, | |
| "loss": 0.4491, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 3.8382904529571533, | |
| "learning_rate": 3.470809668465803e-06, | |
| "loss": 0.4236, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 3.1422176361083984, | |
| "learning_rate": 3.0369584599075772e-06, | |
| "loss": 0.4383, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.356064558029175, | |
| "learning_rate": 2.6031072513493522e-06, | |
| "loss": 0.4434, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 3.1500184535980225, | |
| "learning_rate": 2.1692560427911268e-06, | |
| "loss": 0.3509, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 2.9784765243530273, | |
| "learning_rate": 1.7354048342329016e-06, | |
| "loss": 0.4451, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 8.244532585144043, | |
| "learning_rate": 1.3015536256746761e-06, | |
| "loss": 0.381, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 8.15170669555664, | |
| "learning_rate": 8.677024171164508e-07, | |
| "loss": 0.4256, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 3.7399942874908447, | |
| "learning_rate": 4.338512085582254e-07, | |
| "loss": 0.3909, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.0128397941589355, | |
| "learning_rate": 0.0, | |
| "loss": 0.3897, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7775, | |
| "eval_f1": 0.6307053941908713, | |
| "eval_loss": 0.4450126588344574, | |
| "eval_precision": 0.6440677966101694, | |
| "eval_recall": 0.6178861788617886, | |
| "eval_runtime": 1.5626, | |
| "eval_samples_per_second": 255.985, | |
| "eval_steps_per_second": 15.999, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 423630740901888.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 8.677024171164507e-06, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 16, | |
| "seed": 37 | |
| } | |
| } | |