| { | |
| "best_metric": 0.6813856959342957, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_sst2_256/checkpoint-527", | |
| "epoch": 6.0, | |
| "global_step": 3162, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6662, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5561926605504587, | |
| "eval_loss": 0.6813856959342957, | |
| "eval_runtime": 1.5529, | |
| "eval_samples_per_second": 561.531, | |
| "eval_steps_per_second": 4.508, | |
| "step": 527 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5954, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5493119266055045, | |
| "eval_loss": 0.7089729905128479, | |
| "eval_runtime": 1.5587, | |
| "eval_samples_per_second": 559.425, | |
| "eval_steps_per_second": 4.491, | |
| "step": 1054 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5689, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5596330275229358, | |
| "eval_loss": 0.715036153793335, | |
| "eval_runtime": 1.6372, | |
| "eval_samples_per_second": 532.618, | |
| "eval_steps_per_second": 4.276, | |
| "step": 1581 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5546, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5538990825688074, | |
| "eval_loss": 0.6892502307891846, | |
| "eval_runtime": 1.5477, | |
| "eval_samples_per_second": 563.399, | |
| "eval_steps_per_second": 4.523, | |
| "step": 2108 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5473, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5871559633027523, | |
| "eval_loss": 0.7050522565841675, | |
| "eval_runtime": 1.5988, | |
| "eval_samples_per_second": 545.394, | |
| "eval_steps_per_second": 4.378, | |
| "step": 2635 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5421, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5871559633027523, | |
| "eval_loss": 0.69826340675354, | |
| "eval_runtime": 1.5749, | |
| "eval_samples_per_second": 553.7, | |
| "eval_steps_per_second": 4.445, | |
| "step": 3162 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 3162, | |
| "total_flos": 1.0415609493848064e+16, | |
| "train_loss": 0.5790761504876017, | |
| "train_runtime": 2176.4824, | |
| "train_samples_per_second": 1547.198, | |
| "train_steps_per_second": 12.107 | |
| } | |
| ], | |
| "max_steps": 26350, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.0415609493848064e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |