| { | |
| "best_metric": 0.7438596491228071, | |
| "best_model_checkpoint": "final_models/glue_models/structroberta_s1_final//finetune/qnli/checkpoint-2000", | |
| "epoch": 10.0, | |
| "global_step": 3660, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.6854768395423889, | |
| "eval_f1": 0.698279479647503, | |
| "eval_loss": 0.5918503403663635, | |
| "eval_mcc": 0.37039988909838023, | |
| "eval_runtime": 4.521, | |
| "eval_samples_per_second": 505.638, | |
| "eval_steps_per_second": 63.26, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.316939890710383e-05, | |
| "loss": 0.6348, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_accuracy": 0.6942257285118103, | |
| "eval_f1": 0.7279096924873493, | |
| "eval_loss": 0.6217535138130188, | |
| "eval_mcc": 0.3841755069564845, | |
| "eval_runtime": 4.5007, | |
| "eval_samples_per_second": 507.921, | |
| "eval_steps_per_second": 63.546, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 3.633879781420765e-05, | |
| "loss": 0.5113, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_accuracy": 0.6999124884605408, | |
| "eval_f1": 0.7262569832402235, | |
| "eval_loss": 0.6595348715782166, | |
| "eval_mcc": 0.39577981158604925, | |
| "eval_runtime": 4.4963, | |
| "eval_samples_per_second": 508.422, | |
| "eval_steps_per_second": 63.608, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 2.9508196721311478e-05, | |
| "loss": 0.3571, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "eval_accuracy": 0.7104111909866333, | |
| "eval_f1": 0.7324171382376718, | |
| "eval_loss": 0.8105891942977905, | |
| "eval_mcc": 0.4174093343048448, | |
| "eval_runtime": 4.5019, | |
| "eval_samples_per_second": 507.789, | |
| "eval_steps_per_second": 63.529, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 2.2677595628415303e-05, | |
| "loss": 0.2085, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "eval_accuracy": 0.712598443031311, | |
| "eval_f1": 0.7438596491228071, | |
| "eval_loss": 0.8948610424995422, | |
| "eval_mcc": 0.42158343510176083, | |
| "eval_runtime": 4.5038, | |
| "eval_samples_per_second": 507.569, | |
| "eval_steps_per_second": 63.502, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "eval_accuracy": 0.7213473320007324, | |
| "eval_f1": 0.7357942762339277, | |
| "eval_loss": 1.1155877113342285, | |
| "eval_mcc": 0.44112337210044195, | |
| "eval_runtime": 4.4995, | |
| "eval_samples_per_second": 508.053, | |
| "eval_steps_per_second": 63.562, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 1.5846994535519128e-05, | |
| "loss": 0.1211, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "eval_accuracy": 0.7152230739593506, | |
| "eval_f1": 0.740534077321642, | |
| "eval_loss": 1.3353219032287598, | |
| "eval_mcc": 0.42670434808360297, | |
| "eval_runtime": 4.5055, | |
| "eval_samples_per_second": 507.381, | |
| "eval_steps_per_second": 63.478, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 9.016393442622952e-06, | |
| "loss": 0.0599, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "eval_accuracy": 0.7121610045433044, | |
| "eval_f1": 0.7269709543568464, | |
| "eval_loss": 1.4796608686447144, | |
| "eval_mcc": 0.42273254192250986, | |
| "eval_runtime": 4.5005, | |
| "eval_samples_per_second": 507.939, | |
| "eval_steps_per_second": 63.548, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 2.185792349726776e-06, | |
| "loss": 0.0317, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "eval_accuracy": 0.7108486294746399, | |
| "eval_f1": 0.7300939158840344, | |
| "eval_loss": 1.662826418876648, | |
| "eval_mcc": 0.4188408674974473, | |
| "eval_runtime": 4.4953, | |
| "eval_samples_per_second": 508.53, | |
| "eval_steps_per_second": 63.622, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 3660, | |
| "total_flos": 3.664834789427712e+16, | |
| "train_loss": 0.2638393348032008, | |
| "train_runtime": 2094.1582, | |
| "train_samples_per_second": 209.712, | |
| "train_steps_per_second": 1.748 | |
| } | |
| ], | |
| "max_steps": 3660, | |
| "num_train_epochs": 10, | |
| "total_flos": 3.664834789427712e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |