| { | |
| "best_metric": 1.410552978515625, | |
| "best_model_checkpoint": "./mbartLarge_mid_en-ko1/checkpoint-7500", | |
| "epoch": 8.978675645342312, | |
| "eval_steps": 1500, | |
| "global_step": 12000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1372, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.998899601395987e-05, | |
| "loss": 1.7299, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.995599374285618e-05, | |
| "loss": 1.5855, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_bleu": 11.5186, | |
| "eval_gen_len": 16.204, | |
| "eval_loss": 1.5215433835983276, | |
| "eval_runtime": 165.3535, | |
| "eval_samples_per_second": 16.159, | |
| "eval_steps_per_second": 1.01, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.990102223921136e-05, | |
| "loss": 1.499, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 4.9824129895478125e-05, | |
| "loss": 1.4854, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 4.9725384401438634e-05, | |
| "loss": 1.4287, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_bleu": 12.2855, | |
| "eval_gen_len": 16.1497, | |
| "eval_loss": 1.4549453258514404, | |
| "eval_runtime": 165.6095, | |
| "eval_samples_per_second": 16.134, | |
| "eval_steps_per_second": 1.008, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.960487268461591e-05, | |
| "loss": 1.38, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.946270083374993e-05, | |
| "loss": 1.385, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 4.929899400540567e-05, | |
| "loss": 1.2937, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "eval_bleu": 12.6484, | |
| "eval_gen_len": 16.2152, | |
| "eval_loss": 1.4249523878097534, | |
| "eval_runtime": 169.9417, | |
| "eval_samples_per_second": 15.723, | |
| "eval_steps_per_second": 0.983, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 4.9113896313795424e-05, | |
| "loss": 1.3033, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 4.8907570703912376e-05, | |
| "loss": 1.2837, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 4.868019880808697e-05, | |
| "loss": 1.2444, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "eval_bleu": 13.0063, | |
| "eval_gen_len": 16.0749, | |
| "eval_loss": 1.4164687395095825, | |
| "eval_runtime": 167.1035, | |
| "eval_samples_per_second": 15.99, | |
| "eval_steps_per_second": 0.999, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 4.843198078609264e-05, | |
| "loss": 1.2463, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.816313514894128e-05, | |
| "loss": 1.1724, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 4.787389856652394e-05, | |
| "loss": 1.1335, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "eval_bleu": 13.2758, | |
| "eval_gen_len": 16.235, | |
| "eval_loss": 1.410552978515625, | |
| "eval_runtime": 168.3674, | |
| "eval_samples_per_second": 15.87, | |
| "eval_steps_per_second": 0.992, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 4.7564525659265865e-05, | |
| "loss": 1.1514, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 4.7235288773979234e-05, | |
| "loss": 1.0421, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 4.688647774411124e-05, | |
| "loss": 1.0508, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "eval_bleu": 13.0601, | |
| "eval_gen_len": 15.86, | |
| "eval_loss": 1.4243229627609253, | |
| "eval_runtime": 163.5075, | |
| "eval_samples_per_second": 16.342, | |
| "eval_steps_per_second": 1.021, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 4.651839963459813e-05, | |
| "loss": 1.0106, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 4.6131378471550225e-05, | |
| "loss": 0.9055, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 4.572575495700553e-05, | |
| "loss": 0.9462, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "eval_bleu": 13.0828, | |
| "eval_gen_len": 16.0475, | |
| "eval_loss": 1.4497138261795044, | |
| "eval_runtime": 157.7664, | |
| "eval_samples_per_second": 16.936, | |
| "eval_steps_per_second": 1.059, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 4.53018861690034e-05, | |
| "loss": 0.8717, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 4.48601452472419e-05, | |
| "loss": 0.8452, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 4.4400921064595953e-05, | |
| "loss": 0.8464, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_bleu": 13.5878, | |
| "eval_gen_len": 15.9308, | |
| "eval_loss": 1.4692068099975586, | |
| "eval_runtime": 157.7767, | |
| "eval_samples_per_second": 16.935, | |
| "eval_steps_per_second": 1.058, | |
| "step": 12000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 53440, | |
| "num_train_epochs": 40, | |
| "save_steps": 1500, | |
| "total_flos": 4.162969604383048e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |