| { | |
| "best_metric": 0.5551439523696899, | |
| "best_model_checkpoint": "output/checkpoint-400", | |
| "epoch": 0.6417326782312243, | |
| "eval_steps": 50, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4.444444444444445e-07, | |
| "loss": 1.0919, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7661, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 0.6954500079154968, | |
| "eval_runtime": 28.4403, | |
| "eval_samples_per_second": 6.153, | |
| "eval_steps_per_second": 1.547, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.222222222222222e-05, | |
| "loss": 0.6119, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.7314285714285714, | |
| "eval_loss": 0.5819395780563354, | |
| "eval_runtime": 28.3595, | |
| "eval_samples_per_second": 6.171, | |
| "eval_steps_per_second": 1.552, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 6.444444444444446e-05, | |
| "loss": 0.5869, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.7371428571428571, | |
| "eval_loss": 0.6004045605659485, | |
| "eval_runtime": 28.3043, | |
| "eval_samples_per_second": 6.183, | |
| "eval_steps_per_second": 1.555, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.588, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.76, | |
| "eval_loss": 0.5582989454269409, | |
| "eval_runtime": 28.3002, | |
| "eval_samples_per_second": 6.184, | |
| "eval_steps_per_second": 1.555, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.80411361410382e-05, | |
| "loss": 0.5701, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 0.6456161141395569, | |
| "eval_runtime": 28.2527, | |
| "eval_samples_per_second": 6.194, | |
| "eval_steps_per_second": 1.557, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.31439764936337e-05, | |
| "loss": 0.5652, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.72, | |
| "eval_loss": 0.6802676916122437, | |
| "eval_runtime": 28.3389, | |
| "eval_samples_per_second": 6.175, | |
| "eval_steps_per_second": 1.553, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 8.82468168462292e-05, | |
| "loss": 0.5936, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.7542857142857143, | |
| "eval_loss": 0.5668751001358032, | |
| "eval_runtime": 28.2611, | |
| "eval_samples_per_second": 6.192, | |
| "eval_steps_per_second": 1.557, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8.334965719882468e-05, | |
| "loss": 0.5961, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_accuracy": 0.76, | |
| "eval_loss": 0.5551439523696899, | |
| "eval_runtime": 26.2906, | |
| "eval_samples_per_second": 6.656, | |
| "eval_steps_per_second": 1.674, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 7.845249755142018e-05, | |
| "loss": 0.5842, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.7314285714285714, | |
| "eval_loss": 0.6041642427444458, | |
| "eval_runtime": 26.2867, | |
| "eval_samples_per_second": 6.657, | |
| "eval_steps_per_second": 1.674, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 7.355533790401568e-05, | |
| "loss": 0.5749, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.7257142857142858, | |
| "eval_loss": 0.6361325979232788, | |
| "eval_runtime": 28.3165, | |
| "eval_samples_per_second": 6.18, | |
| "eval_steps_per_second": 1.554, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 6.865817825661118e-05, | |
| "loss": 0.6174, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.7485714285714286, | |
| "eval_loss": 0.582287073135376, | |
| "eval_runtime": 28.3412, | |
| "eval_samples_per_second": 6.175, | |
| "eval_steps_per_second": 1.553, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.376101860920667e-05, | |
| "loss": 0.5752, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.7314285714285714, | |
| "eval_loss": 0.6466975212097168, | |
| "eval_runtime": 28.3593, | |
| "eval_samples_per_second": 6.171, | |
| "eval_steps_per_second": 1.552, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 5.886385896180215e-05, | |
| "loss": 0.5679, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_accuracy": 0.72, | |
| "eval_loss": 0.5773769617080688, | |
| "eval_runtime": 28.2118, | |
| "eval_samples_per_second": 6.203, | |
| "eval_steps_per_second": 1.56, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 5.396669931439765e-05, | |
| "loss": 0.5793, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.6914285714285714, | |
| "eval_loss": 0.5711085796356201, | |
| "eval_runtime": 28.2951, | |
| "eval_samples_per_second": 6.185, | |
| "eval_steps_per_second": 1.555, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.906953966699315e-05, | |
| "loss": 0.5897, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.7028571428571428, | |
| "eval_loss": 0.5825155377388, | |
| "eval_runtime": 28.19, | |
| "eval_samples_per_second": 6.208, | |
| "eval_steps_per_second": 1.561, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.4172380019588636e-05, | |
| "loss": 0.5782, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_accuracy": 0.7257142857142858, | |
| "eval_loss": 0.5976516008377075, | |
| "eval_runtime": 28.2521, | |
| "eval_samples_per_second": 6.194, | |
| "eval_steps_per_second": 1.557, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1246, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "total_flos": 2.963160528039936e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |