| { | |
| "best_metric": 1.0586252212524414, | |
| "best_model_checkpoint": "distilbert_sa_GLUE_Experiment_data_aug_qnli_256/checkpoint-16604", | |
| "epoch": 6.0, | |
| "global_step": 99624, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9000240905805834e-05, | |
| "loss": 0.4745, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5680029287937031, | |
| "eval_loss": 1.0586252212524414, | |
| "eval_runtime": 2.0901, | |
| "eval_samples_per_second": 2613.768, | |
| "eval_steps_per_second": 10.526, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.800066249096604e-05, | |
| "loss": 0.2251, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5707486728903532, | |
| "eval_loss": 1.3085120916366577, | |
| "eval_runtime": 2.0886, | |
| "eval_samples_per_second": 2615.588, | |
| "eval_steps_per_second": 10.533, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.700114430257769e-05, | |
| "loss": 0.1318, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5808164012447373, | |
| "eval_loss": 1.4266568422317505, | |
| "eval_runtime": 2.1547, | |
| "eval_samples_per_second": 2535.444, | |
| "eval_steps_per_second": 10.21, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600144543483498e-05, | |
| "loss": 0.0838, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5778876075416438, | |
| "eval_loss": 1.6666687726974487, | |
| "eval_runtime": 2.0958, | |
| "eval_samples_per_second": 2606.607, | |
| "eval_steps_per_second": 10.497, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5001867019995185e-05, | |
| "loss": 0.0572, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.576423210690097, | |
| "eval_loss": 1.9572008848190308, | |
| "eval_runtime": 2.0878, | |
| "eval_samples_per_second": 2616.615, | |
| "eval_steps_per_second": 10.537, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400228860515538e-05, | |
| "loss": 0.041, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5744096650192202, | |
| "eval_loss": 2.0601627826690674, | |
| "eval_runtime": 2.0721, | |
| "eval_samples_per_second": 2636.402, | |
| "eval_steps_per_second": 10.617, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 99624, | |
| "total_flos": 4.3517441866727424e+17, | |
| "train_loss": 0.16891067297805373, | |
| "train_runtime": 17984.6912, | |
| "train_samples_per_second": 11817.025, | |
| "train_steps_per_second": 46.161 | |
| } | |
| ], | |
| "max_steps": 830200, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.3517441866727424e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |