| { | |
| "best_metric": 0.6392325758934021, | |
| "best_model_checkpoint": "output/checkpoint-700", | |
| "epoch": 0.8823824325679335, | |
| "eval_steps": 50, | |
| "global_step": 1100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.3333333333333332e-06, | |
| "loss": 0.785, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 0.6547, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.6626865671641791, | |
| "eval_loss": 0.8300915956497192, | |
| "eval_runtime": 61.5689, | |
| "eval_samples_per_second": 5.441, | |
| "eval_steps_per_second": 1.364, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00012666666666666666, | |
| "loss": 0.6654, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.6895522388059702, | |
| "eval_loss": 0.6663276553153992, | |
| "eval_runtime": 56.4633, | |
| "eval_samples_per_second": 5.933, | |
| "eval_steps_per_second": 1.488, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019333333333333333, | |
| "loss": 0.6491, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.6537313432835821, | |
| "eval_loss": 0.9172552824020386, | |
| "eval_runtime": 56.4705, | |
| "eval_samples_per_second": 5.932, | |
| "eval_steps_per_second": 1.488, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00026, | |
| "loss": 0.641, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.6865671641791045, | |
| "eval_loss": 0.6528732180595398, | |
| "eval_runtime": 56.4485, | |
| "eval_samples_per_second": 5.935, | |
| "eval_steps_per_second": 1.488, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0002941234084231146, | |
| "loss": 0.6521, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.6835820895522388, | |
| "eval_loss": 0.8197052478790283, | |
| "eval_runtime": 56.4577, | |
| "eval_samples_per_second": 5.934, | |
| "eval_steps_per_second": 1.488, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00027943192948090106, | |
| "loss": 0.638, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.6835820895522388, | |
| "eval_loss": 0.6961421370506287, | |
| "eval_runtime": 56.4506, | |
| "eval_samples_per_second": 5.934, | |
| "eval_steps_per_second": 1.488, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00026474045053868754, | |
| "loss": 0.6857, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.6835820895522388, | |
| "eval_loss": 0.7876304388046265, | |
| "eval_runtime": 56.4651, | |
| "eval_samples_per_second": 5.933, | |
| "eval_steps_per_second": 1.488, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.000250048971596474, | |
| "loss": 0.6469, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_accuracy": 0.6626865671641791, | |
| "eval_loss": 0.7487305402755737, | |
| "eval_runtime": 56.4489, | |
| "eval_samples_per_second": 5.935, | |
| "eval_steps_per_second": 1.488, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00023535749265426052, | |
| "loss": 0.6813, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.6626865671641791, | |
| "eval_loss": 0.8447557091712952, | |
| "eval_runtime": 56.4428, | |
| "eval_samples_per_second": 5.935, | |
| "eval_steps_per_second": 1.488, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.000220666013712047, | |
| "loss": 0.645, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.6835820895522388, | |
| "eval_loss": 0.6624641418457031, | |
| "eval_runtime": 56.4531, | |
| "eval_samples_per_second": 5.934, | |
| "eval_steps_per_second": 1.488, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002059745347698335, | |
| "loss": 0.7044, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_accuracy": 0.7014925373134329, | |
| "eval_loss": 0.63779217004776, | |
| "eval_runtime": 56.4382, | |
| "eval_samples_per_second": 5.936, | |
| "eval_steps_per_second": 1.488, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00019128305582761997, | |
| "loss": 0.6399, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.6716417910447762, | |
| "eval_loss": 0.7784742712974548, | |
| "eval_runtime": 56.4356, | |
| "eval_samples_per_second": 5.936, | |
| "eval_steps_per_second": 1.488, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00017659157688540645, | |
| "loss": 0.6039, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_accuracy": 0.6985074626865672, | |
| "eval_loss": 0.665508508682251, | |
| "eval_runtime": 56.4424, | |
| "eval_samples_per_second": 5.935, | |
| "eval_steps_per_second": 1.488, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00016190009794319292, | |
| "loss": 0.6321, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.6716417910447762, | |
| "eval_loss": 0.6392325758934021, | |
| "eval_runtime": 56.437, | |
| "eval_samples_per_second": 5.936, | |
| "eval_steps_per_second": 1.488, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00014720861900097942, | |
| "loss": 0.6448, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.6686567164179105, | |
| "eval_loss": 0.7872520089149475, | |
| "eval_runtime": 56.4423, | |
| "eval_samples_per_second": 5.935, | |
| "eval_steps_per_second": 1.488, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001325171400587659, | |
| "loss": 0.6033, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_accuracy": 0.6895522388059702, | |
| "eval_loss": 0.723906934261322, | |
| "eval_runtime": 56.4576, | |
| "eval_samples_per_second": 5.934, | |
| "eval_steps_per_second": 1.488, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00011782566111655239, | |
| "loss": 0.6174, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.6925373134328359, | |
| "eval_loss": 0.6893543601036072, | |
| "eval_runtime": 56.4403, | |
| "eval_samples_per_second": 5.935, | |
| "eval_steps_per_second": 1.488, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00010313418217433888, | |
| "loss": 0.5859, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.6746268656716418, | |
| "eval_loss": 0.6730663776397705, | |
| "eval_runtime": 56.4424, | |
| "eval_samples_per_second": 5.935, | |
| "eval_steps_per_second": 1.488, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.844270323212536e-05, | |
| "loss": 0.6, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.6686567164179105, | |
| "eval_loss": 0.6526519060134888, | |
| "eval_runtime": 56.4284, | |
| "eval_samples_per_second": 5.937, | |
| "eval_steps_per_second": 1.489, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.375122428991184e-05, | |
| "loss": 0.6127, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.7014925373134329, | |
| "eval_loss": 0.6604557037353516, | |
| "eval_runtime": 56.4364, | |
| "eval_samples_per_second": 5.936, | |
| "eval_steps_per_second": 1.488, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 5.9059745347698334e-05, | |
| "loss": 0.6186, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.6865671641791045, | |
| "eval_loss": 0.6518017053604126, | |
| "eval_runtime": 56.4388, | |
| "eval_samples_per_second": 5.936, | |
| "eval_steps_per_second": 1.488, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.4368266405484816e-05, | |
| "loss": 0.5929, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_accuracy": 0.6985074626865672, | |
| "eval_loss": 0.6976536512374878, | |
| "eval_runtime": 56.4535, | |
| "eval_samples_per_second": 5.934, | |
| "eval_steps_per_second": 1.488, | |
| "step": 1100 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1246, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 4.1300552228438016e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |