| { | |
| "best_metric": 0.6648005843162537, | |
| "best_model_checkpoint": "distilbert_add_GLUE_Experiment_qnli/checkpoint-410", | |
| "epoch": 6.0, | |
| "global_step": 2460, | |
| "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": 0.6886, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6066263957532492, | |
| "eval_loss": 0.6648005843162537, | |
| "eval_runtime": 4.3884, | |
| "eval_samples_per_second": 1244.864, | |
| "eval_steps_per_second": 5.013, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6569, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5998535603148453, | |
| "eval_loss": 0.6677054762840271, | |
| "eval_runtime": 4.3835, | |
| "eval_samples_per_second": 1246.26, | |
| "eval_steps_per_second": 5.019, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6419, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5914332784184514, | |
| "eval_loss": 0.6671571731567383, | |
| "eval_runtime": 4.383, | |
| "eval_samples_per_second": 1246.413, | |
| "eval_steps_per_second": 5.019, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6293, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5976569650375252, | |
| "eval_loss": 0.6676780581474304, | |
| "eval_runtime": 4.085, | |
| "eval_samples_per_second": 1337.341, | |
| "eval_steps_per_second": 5.386, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6118, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.600219659527732, | |
| "eval_loss": 0.6690582633018494, | |
| "eval_runtime": 4.3841, | |
| "eval_samples_per_second": 1246.106, | |
| "eval_steps_per_second": 5.018, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5857, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6077246933919092, | |
| "eval_loss": 0.6853577494621277, | |
| "eval_runtime": 4.3892, | |
| "eval_samples_per_second": 1244.634, | |
| "eval_steps_per_second": 5.012, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 2460, | |
| "total_flos": 3.831136168339046e+16, | |
| "train_loss": 0.6356980114448362, | |
| "train_runtime": 1226.2807, | |
| "train_samples_per_second": 4270.759, | |
| "train_steps_per_second": 16.717 | |
| } | |
| ], | |
| "max_steps": 20500, | |
| "num_train_epochs": 50, | |
| "total_flos": 3.831136168339046e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |