| { | |
| "best_metric": 0.665068507194519, | |
| "best_model_checkpoint": "output/checkpoint-50", | |
| "epoch": 2.455242966751918, | |
| "eval_steps": 50, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 7.575757575757576e-07, | |
| "loss": 0.4968, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3.484848484848485e-05, | |
| "loss": 0.5254, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.6507462686567164, | |
| "eval_loss": 0.665068507194519, | |
| "eval_runtime": 61.5592, | |
| "eval_samples_per_second": 5.442, | |
| "eval_steps_per_second": 1.365, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.4971, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_accuracy": 0.6656716417910448, | |
| "eval_loss": 0.8002150058746338, | |
| "eval_runtime": 61.211, | |
| "eval_samples_per_second": 5.473, | |
| "eval_steps_per_second": 1.372, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 9.783333333333334e-05, | |
| "loss": 0.5039, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.6865671641791045, | |
| "eval_loss": 0.7404947280883789, | |
| "eval_runtime": 61.4487, | |
| "eval_samples_per_second": 5.452, | |
| "eval_steps_per_second": 1.367, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 0.4944, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_accuracy": 0.6238805970149254, | |
| "eval_loss": 0.9986834526062012, | |
| "eval_runtime": 60.5135, | |
| "eval_samples_per_second": 5.536, | |
| "eval_steps_per_second": 1.388, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 8.116666666666667e-05, | |
| "loss": 0.508, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.6268656716417911, | |
| "eval_loss": 0.9990125894546509, | |
| "eval_runtime": 56.4381, | |
| "eval_samples_per_second": 5.936, | |
| "eval_steps_per_second": 1.488, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.283333333333335e-05, | |
| "loss": 0.4904, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_accuracy": 0.6119402985074627, | |
| "eval_loss": 1.0483125448226929, | |
| "eval_runtime": 56.4117, | |
| "eval_samples_per_second": 5.938, | |
| "eval_steps_per_second": 1.489, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 0.4694, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_accuracy": 0.6417910447761194, | |
| "eval_loss": 0.9735069274902344, | |
| "eval_runtime": 56.422, | |
| "eval_samples_per_second": 5.937, | |
| "eval_steps_per_second": 1.489, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 5.6166666666666665e-05, | |
| "loss": 0.4937, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_accuracy": 0.5880597014925373, | |
| "eval_loss": 1.1042370796203613, | |
| "eval_runtime": 56.4248, | |
| "eval_samples_per_second": 5.937, | |
| "eval_steps_per_second": 1.489, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.7833333333333335e-05, | |
| "loss": 0.4877, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_accuracy": 0.6477611940298508, | |
| "eval_loss": 0.9591583609580994, | |
| "eval_runtime": 56.4028, | |
| "eval_samples_per_second": 5.939, | |
| "eval_steps_per_second": 1.489, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.482, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_accuracy": 0.6716417910447762, | |
| "eval_loss": 0.89241623878479, | |
| "eval_runtime": 56.408, | |
| "eval_samples_per_second": 5.939, | |
| "eval_steps_per_second": 1.489, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 3.116666666666667e-05, | |
| "loss": 0.4688, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_accuracy": 0.6447761194029851, | |
| "eval_loss": 0.9692543148994446, | |
| "eval_runtime": 56.4147, | |
| "eval_samples_per_second": 5.938, | |
| "eval_steps_per_second": 1.489, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 2.2833333333333334e-05, | |
| "loss": 0.4818, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "eval_accuracy": 0.6417910447761194, | |
| "eval_loss": 0.981917679309845, | |
| "eval_runtime": 56.4148, | |
| "eval_samples_per_second": 5.938, | |
| "eval_steps_per_second": 1.489, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 732, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "total_flos": 2.5561235391596544e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |