| { | |
| "best_metric": 1.146978735923767, | |
| "best_model_checkpoint": "./jako_mbartLarge_6p/checkpoint-4000", | |
| "epoch": 3.838771593090211, | |
| "eval_steps": 1000, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.951314508276534e-05, | |
| "loss": 1.88, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.829600778967868e-05, | |
| "loss": 1.4498, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_bleu": 16.7947, | |
| "eval_gen_len": 20.0778, | |
| "eval_loss": 1.3302855491638184, | |
| "eval_runtime": 299.3898, | |
| "eval_samples_per_second": 13.915, | |
| "eval_steps_per_second": 0.872, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.707887049659202e-05, | |
| "loss": 1.3412, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.5861733203505356e-05, | |
| "loss": 1.263, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_bleu": 18.6214, | |
| "eval_gen_len": 20.6673, | |
| "eval_loss": 1.1883659362792969, | |
| "eval_runtime": 336.2683, | |
| "eval_samples_per_second": 12.389, | |
| "eval_steps_per_second": 0.776, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.4644595910418695e-05, | |
| "loss": 1.0852, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 4.342745861733204e-05, | |
| "loss": 0.8999, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_bleu": 18.9014, | |
| "eval_gen_len": 19.4064, | |
| "eval_loss": 1.1701579093933105, | |
| "eval_runtime": 290.3573, | |
| "eval_samples_per_second": 14.348, | |
| "eval_steps_per_second": 0.899, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.221032132424538e-05, | |
| "loss": 0.8646, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 4.099318403115872e-05, | |
| "loss": 0.827, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_bleu": 21.2511, | |
| "eval_gen_len": 19.566, | |
| "eval_loss": 1.146978735923767, | |
| "eval_runtime": 287.6889, | |
| "eval_samples_per_second": 14.481, | |
| "eval_steps_per_second": 0.907, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 3.977604673807206e-05, | |
| "loss": 0.7493, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 3.85589094449854e-05, | |
| "loss": 0.5948, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_bleu": 20.5498, | |
| "eval_gen_len": 19.3941, | |
| "eval_loss": 1.2114742994308472, | |
| "eval_runtime": 292.2098, | |
| "eval_samples_per_second": 14.257, | |
| "eval_steps_per_second": 0.893, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 3.7341772151898736e-05, | |
| "loss": 0.5631, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 3.6124634858812075e-05, | |
| "loss": 0.5404, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_bleu": 21.2187, | |
| "eval_gen_len": 19.4357, | |
| "eval_loss": 1.2166906595230103, | |
| "eval_runtime": 288.5479, | |
| "eval_samples_per_second": 14.438, | |
| "eval_steps_per_second": 0.905, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 3.4907497565725414e-05, | |
| "loss": 0.5082, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.369036027263875e-05, | |
| "loss": 0.3842, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_bleu": 20.7732, | |
| "eval_gen_len": 19.3493, | |
| "eval_loss": 1.271054983139038, | |
| "eval_runtime": 289.0747, | |
| "eval_samples_per_second": 14.411, | |
| "eval_steps_per_second": 0.903, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 3.247322297955209e-05, | |
| "loss": 0.3553, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 3.125608568646544e-05, | |
| "loss": 0.3419, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_bleu": 20.7001, | |
| "eval_gen_len": 19.3106, | |
| "eval_loss": 1.290993571281433, | |
| "eval_runtime": 283.9873, | |
| "eval_samples_per_second": 14.67, | |
| "eval_steps_per_second": 0.919, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "step": 8000, | |
| "total_flos": 2.77392610295808e+17, | |
| "train_loss": 0.8529902515411377, | |
| "train_runtime": 11046.7228, | |
| "train_samples_per_second": 30.175, | |
| "train_steps_per_second": 1.887 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20840, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000, | |
| "total_flos": 2.77392610295808e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |