| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.776119402985074, | |
| "eval_steps": 200, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.4891, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_accuracy": 0.7833955223880597, | |
| "eval_loss": 0.4603855013847351, | |
| "eval_runtime": 253.4527, | |
| "eval_samples_per_second": 63.444, | |
| "eval_steps_per_second": 10.574, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4781, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.7914179104477612, | |
| "eval_loss": 0.44968461990356445, | |
| "eval_runtime": 253.7178, | |
| "eval_samples_per_second": 63.378, | |
| "eval_steps_per_second": 10.563, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.722222222222222e-06, | |
| "loss": 0.4697, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.7952114427860697, | |
| "eval_loss": 0.43988558650016785, | |
| "eval_runtime": 253.4986, | |
| "eval_samples_per_second": 63.432, | |
| "eval_steps_per_second": 10.572, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.459, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_accuracy": 0.8071517412935323, | |
| "eval_loss": 0.4225695729255676, | |
| "eval_runtime": 256.265, | |
| "eval_samples_per_second": 62.748, | |
| "eval_steps_per_second": 10.458, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.4369, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_accuracy": 0.8149875621890548, | |
| "eval_loss": 0.4131614863872528, | |
| "eval_runtime": 254.6359, | |
| "eval_samples_per_second": 63.149, | |
| "eval_steps_per_second": 10.525, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 3.88888888888889e-06, | |
| "loss": 0.4232, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_accuracy": 0.818594527363184, | |
| "eval_loss": 0.4021635055541992, | |
| "eval_runtime": 256.0534, | |
| "eval_samples_per_second": 62.799, | |
| "eval_steps_per_second": 10.467, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 3.6111111111111115e-06, | |
| "loss": 0.4234, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_accuracy": 0.8263059701492538, | |
| "eval_loss": 0.3889436423778534, | |
| "eval_runtime": 253.8628, | |
| "eval_samples_per_second": 63.341, | |
| "eval_steps_per_second": 10.557, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.4016, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 0.8316542288557214, | |
| "eval_loss": 0.3784995973110199, | |
| "eval_runtime": 253.3618, | |
| "eval_samples_per_second": 63.467, | |
| "eval_steps_per_second": 10.578, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 3.055555555555556e-06, | |
| "loss": 0.3853, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.8365671641791045, | |
| "eval_loss": 0.3704816401004791, | |
| "eval_runtime": 253.6776, | |
| "eval_samples_per_second": 63.388, | |
| "eval_steps_per_second": 10.565, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "loss": 0.3777, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_accuracy": 0.8468905472636816, | |
| "eval_loss": 0.35585275292396545, | |
| "eval_runtime": 279.3911, | |
| "eval_samples_per_second": 57.554, | |
| "eval_steps_per_second": 9.592, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.3681, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_accuracy": 0.8519900497512438, | |
| "eval_loss": 0.34731847047805786, | |
| "eval_runtime": 257.1097, | |
| "eval_samples_per_second": 62.541, | |
| "eval_steps_per_second": 10.424, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 2.222222222222222e-06, | |
| "loss": 0.3542, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "eval_accuracy": 0.8529228855721394, | |
| "eval_loss": 0.3429964780807495, | |
| "eval_runtime": 261.8479, | |
| "eval_samples_per_second": 61.41, | |
| "eval_steps_per_second": 10.235, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 1.944444444444445e-06, | |
| "loss": 0.3445, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_accuracy": 0.8588930348258706, | |
| "eval_loss": 0.3309912085533142, | |
| "eval_runtime": 261.3944, | |
| "eval_samples_per_second": 61.516, | |
| "eval_steps_per_second": 10.253, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.3357, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "eval_accuracy": 0.8645522388059701, | |
| "eval_loss": 0.3250305950641632, | |
| "eval_runtime": 254.0587, | |
| "eval_samples_per_second": 63.292, | |
| "eval_steps_per_second": 10.549, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 1.3888888888888892e-06, | |
| "loss": 0.3265, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "eval_accuracy": 0.8647388059701493, | |
| "eval_loss": 0.3216043710708618, | |
| "eval_runtime": 256.6489, | |
| "eval_samples_per_second": 62.654, | |
| "eval_steps_per_second": 10.442, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.111111111111111e-06, | |
| "loss": 0.3126, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_accuracy": 0.8699004975124378, | |
| "eval_loss": 0.3159268796443939, | |
| "eval_runtime": 258.1285, | |
| "eval_samples_per_second": 62.295, | |
| "eval_steps_per_second": 10.382, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 0.313, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "eval_accuracy": 0.8715174129353234, | |
| "eval_loss": 0.3117404282093048, | |
| "eval_runtime": 256.5391, | |
| "eval_samples_per_second": 62.68, | |
| "eval_steps_per_second": 10.447, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 5.555555555555555e-07, | |
| "loss": 0.3122, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_accuracy": 0.8740049751243781, | |
| "eval_loss": 0.3079596161842346, | |
| "eval_runtime": 255.3334, | |
| "eval_samples_per_second": 62.976, | |
| "eval_steps_per_second": 10.496, | |
| "step": 3600 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 4000, | |
| "num_train_epochs": 6, | |
| "save_steps": 200, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |