| { | |
| "best_metric": 0.44338709115982056, | |
| "best_model_checkpoint": "hBERTv1_no_pretrain_sst2/checkpoint-702", | |
| "epoch": 6.0, | |
| "global_step": 4212, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 0.4323, | |
| "step": 702 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8107798165137615, | |
| "eval_loss": 0.44338709115982056, | |
| "eval_runtime": 4.0911, | |
| "eval_samples_per_second": 213.147, | |
| "eval_steps_per_second": 2.444, | |
| "step": 702 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.2664, | |
| "step": 1404 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.801605504587156, | |
| "eval_loss": 0.5413224697113037, | |
| "eval_runtime": 4.0747, | |
| "eval_samples_per_second": 214.002, | |
| "eval_steps_per_second": 2.454, | |
| "step": 1404 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.2222, | |
| "step": 2106 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8130733944954128, | |
| "eval_loss": 0.5243069529533386, | |
| "eval_runtime": 4.0727, | |
| "eval_samples_per_second": 214.11, | |
| "eval_steps_per_second": 2.455, | |
| "step": 2106 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.2092, | |
| "step": 2808 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8004587155963303, | |
| "eval_loss": 0.6012656688690186, | |
| "eval_runtime": 4.0943, | |
| "eval_samples_per_second": 212.979, | |
| "eval_steps_per_second": 2.442, | |
| "step": 2808 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.2346, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8027522935779816, | |
| "eval_loss": 0.4991849958896637, | |
| "eval_runtime": 4.0513, | |
| "eval_samples_per_second": 215.238, | |
| "eval_steps_per_second": 2.468, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.2444, | |
| "step": 4212 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8004587155963303, | |
| "eval_loss": 0.5316653251647949, | |
| "eval_runtime": 4.0478, | |
| "eval_samples_per_second": 215.428, | |
| "eval_steps_per_second": 2.471, | |
| "step": 4212 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 4212, | |
| "total_flos": 5.984501559302554e+16, | |
| "train_loss": 0.26820123931508005, | |
| "train_runtime": 5787.7575, | |
| "train_samples_per_second": 581.823, | |
| "train_steps_per_second": 6.065 | |
| } | |
| ], | |
| "max_steps": 35100, | |
| "num_train_epochs": 50, | |
| "total_flos": 5.984501559302554e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |