| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.999384615384617, | |
| "global_step": 8120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8820000290870667, | |
| "eval_loss": 0.31410130858421326, | |
| "eval_runtime": 6.7598, | |
| "eval_samples_per_second": 73.967, | |
| "eval_steps_per_second": 9.32, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1.3729229285650701e-05, | |
| "loss": 0.352, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9419999718666077, | |
| "eval_loss": 0.19050274789333344, | |
| "eval_runtime": 6.9043, | |
| "eval_samples_per_second": 72.419, | |
| "eval_steps_per_second": 9.125, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.2828361222287796e-05, | |
| "loss": 0.2057, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.949999988079071, | |
| "eval_loss": 0.21147817373275757, | |
| "eval_runtime": 6.8333, | |
| "eval_samples_per_second": 73.171, | |
| "eval_steps_per_second": 9.219, | |
| "step": 1218 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.1927493158924887e-05, | |
| "loss": 0.1434, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9599999785423279, | |
| "eval_loss": 0.18365992605686188, | |
| "eval_runtime": 6.8687, | |
| "eval_samples_per_second": 72.794, | |
| "eval_steps_per_second": 9.172, | |
| "step": 1624 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1.102662509556198e-05, | |
| "loss": 0.1003, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9520000219345093, | |
| "eval_loss": 0.21692131459712982, | |
| "eval_runtime": 6.8941, | |
| "eval_samples_per_second": 72.526, | |
| "eval_steps_per_second": 9.138, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9459999799728394, | |
| "eval_loss": 0.2830177843570709, | |
| "eval_runtime": 6.8528, | |
| "eval_samples_per_second": 72.963, | |
| "eval_steps_per_second": 9.193, | |
| "step": 2436 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 1.0125757032199073e-05, | |
| "loss": 0.0641, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9459999799728394, | |
| "eval_loss": 0.34864214062690735, | |
| "eval_runtime": 6.912, | |
| "eval_samples_per_second": 72.338, | |
| "eval_steps_per_second": 9.115, | |
| "step": 2842 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 9.224888968836167e-06, | |
| "loss": 0.0371, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9380000233650208, | |
| "eval_loss": 0.4117521345615387, | |
| "eval_runtime": 6.9229, | |
| "eval_samples_per_second": 72.225, | |
| "eval_steps_per_second": 9.1, | |
| "step": 3248 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 8.32402090547326e-06, | |
| "loss": 0.0403, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.26648765802383423, | |
| "eval_runtime": 6.9095, | |
| "eval_samples_per_second": 72.364, | |
| "eval_steps_per_second": 9.118, | |
| "step": 3654 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 7.423152842110353e-06, | |
| "loss": 0.0302, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9480000138282776, | |
| "eval_loss": 0.36225178837776184, | |
| "eval_runtime": 6.8924, | |
| "eval_samples_per_second": 72.544, | |
| "eval_steps_per_second": 9.141, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.3049890100955963, | |
| "eval_runtime": 6.9005, | |
| "eval_samples_per_second": 72.459, | |
| "eval_steps_per_second": 9.13, | |
| "step": 4466 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 6.522284778747446e-06, | |
| "loss": 0.0204, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9559999704360962, | |
| "eval_loss": 0.3233047425746918, | |
| "eval_runtime": 6.9225, | |
| "eval_samples_per_second": 72.229, | |
| "eval_steps_per_second": 9.101, | |
| "step": 4872 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 5.621416715384538e-06, | |
| "loss": 0.0172, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9559999704360962, | |
| "eval_loss": 0.3603793680667877, | |
| "eval_runtime": 6.9127, | |
| "eval_samples_per_second": 72.33, | |
| "eval_steps_per_second": 9.114, | |
| "step": 5278 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 4.720548652021632e-06, | |
| "loss": 0.0106, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9520000219345093, | |
| "eval_loss": 0.37912440299987793, | |
| "eval_runtime": 6.9068, | |
| "eval_samples_per_second": 72.392, | |
| "eval_steps_per_second": 9.121, | |
| "step": 5684 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 3.819680588658725e-06, | |
| "loss": 0.0062, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9559999704360962, | |
| "eval_loss": 0.37980133295059204, | |
| "eval_runtime": 6.9114, | |
| "eval_samples_per_second": 72.344, | |
| "eval_steps_per_second": 9.115, | |
| "step": 6090 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9559999704360962, | |
| "eval_loss": 0.3834003210067749, | |
| "eval_runtime": 6.8795, | |
| "eval_samples_per_second": 72.68, | |
| "eval_steps_per_second": 9.158, | |
| "step": 6496 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 2.9188125252958186e-06, | |
| "loss": 0.0085, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9599999785423279, | |
| "eval_loss": 0.36390435695648193, | |
| "eval_runtime": 6.9114, | |
| "eval_samples_per_second": 72.344, | |
| "eval_steps_per_second": 9.115, | |
| "step": 6902 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.0179444619329115e-06, | |
| "loss": 0.0044, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.36908820271492004, | |
| "eval_runtime": 6.9068, | |
| "eval_samples_per_second": 72.392, | |
| "eval_steps_per_second": 9.121, | |
| "step": 7308 | |
| }, | |
| { | |
| "epoch": 18.47, | |
| "learning_rate": 1.1170763985700047e-06, | |
| "loss": 0.0032, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.3803035318851471, | |
| "eval_runtime": 6.7941, | |
| "eval_samples_per_second": 73.593, | |
| "eval_steps_per_second": 9.273, | |
| "step": 7714 | |
| }, | |
| { | |
| "epoch": 19.7, | |
| "learning_rate": 2.1620833520709765e-07, | |
| "loss": 0.003, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9580000042915344, | |
| "eval_loss": 0.3828106224536896, | |
| "eval_runtime": 6.8168, | |
| "eval_samples_per_second": 73.348, | |
| "eval_steps_per_second": 9.242, | |
| "step": 8120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 8120, | |
| "total_flos": 3.028683737679667e+16, | |
| "train_loss": 0.06455737670213718, | |
| "train_runtime": 7825.8738, | |
| "train_samples_per_second": 16.612, | |
| "train_steps_per_second": 1.038 | |
| } | |
| ], | |
| "max_steps": 8120, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.028683737679667e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |