| { | |
| "best_metric": 0.8238747553816047, | |
| "best_model_checkpoint": "tiny-bert-sst2-distilled/run-6/checkpoint-576", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 576, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7334731817245483, | |
| "learning_rate": 0.00019089508944155825, | |
| "loss": 0.4515, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.761252446183953, | |
| "eval_f1": 0.7676190476190476, | |
| "eval_loss": 0.40420547127723694, | |
| "eval_precision": 0.7476808905380334, | |
| "eval_recall": 0.7886497064579256, | |
| "eval_runtime": 28.5902, | |
| "eval_samples_per_second": 35.747, | |
| "eval_steps_per_second": 1.119, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.9262505769729614, | |
| "learning_rate": 0.00015907924120129856, | |
| "loss": 0.3979, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7925636007827789, | |
| "eval_f1": 0.800376647834275, | |
| "eval_loss": 0.38181060552597046, | |
| "eval_precision": 0.7713248638838476, | |
| "eval_recall": 0.8317025440313112, | |
| "eval_runtime": 28.4377, | |
| "eval_samples_per_second": 35.938, | |
| "eval_steps_per_second": 1.125, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.541796922683716, | |
| "learning_rate": 0.00012726339296103884, | |
| "loss": 0.382, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8199608610567515, | |
| "eval_f1": 0.8302583025830259, | |
| "eval_loss": 0.37195342779159546, | |
| "eval_precision": 0.7853403141361257, | |
| "eval_recall": 0.8806262230919765, | |
| "eval_runtime": 28.3641, | |
| "eval_samples_per_second": 36.031, | |
| "eval_steps_per_second": 1.128, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.71069073677063, | |
| "learning_rate": 9.544754472077913e-05, | |
| "loss": 0.3699, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8091976516634051, | |
| "eval_f1": 0.8086359175662414, | |
| "eval_loss": 0.3656693994998932, | |
| "eval_precision": 0.8110236220472441, | |
| "eval_recall": 0.8062622309197651, | |
| "eval_runtime": 28.049, | |
| "eval_samples_per_second": 36.436, | |
| "eval_steps_per_second": 1.141, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.157362461090088, | |
| "learning_rate": 6.363169648051942e-05, | |
| "loss": 0.364, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8189823874755382, | |
| "eval_f1": 0.8282265552460539, | |
| "eval_loss": 0.3648647964000702, | |
| "eval_precision": 0.7879858657243817, | |
| "eval_recall": 0.87279843444227, | |
| "eval_runtime": 28.1306, | |
| "eval_samples_per_second": 36.331, | |
| "eval_steps_per_second": 1.138, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.1236488819122314, | |
| "learning_rate": 3.181584824025971e-05, | |
| "loss": 0.3606, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8238747553816047, | |
| "eval_f1": 0.83271375464684, | |
| "eval_loss": 0.36466673016548157, | |
| "eval_precision": 0.7929203539823009, | |
| "eval_recall": 0.8767123287671232, | |
| "eval_runtime": 28.0309, | |
| "eval_samples_per_second": 36.46, | |
| "eval_steps_per_second": 1.142, | |
| "step": 576 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 672, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 1414171183680.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.6366473529685639, | |
| "learning_rate": 0.00022271093768181798, | |
| "num_train_epochs": 7, | |
| "temperature": 14 | |
| } | |
| } | |