| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.910988336402701, | |
| "eval_steps": 500, | |
| "global_step": 32000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0009846531614487416, | |
| "loss": 0.5762, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0009693063228974831, | |
| "loss": 0.6042, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0009539594843462247, | |
| "loss": 0.6024, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0009386126457949663, | |
| "loss": 0.6057, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0009232658072437078, | |
| "loss": 0.6143, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0009079189686924494, | |
| "loss": 0.6081, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.000892572130141191, | |
| "loss": 0.611, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.0008772252915899325, | |
| "loss": 0.6104, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0008619091467157766, | |
| "loss": 0.6154, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0008465623081645181, | |
| "loss": 0.6151, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 0.0008312154696132597, | |
| "loss": 0.611, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0008158686310620012, | |
| "loss": 0.617, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0008005524861878454, | |
| "loss": 0.6089, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 3.6822, | |
| "eval_gen_len": 18.235, | |
| "eval_loss": 0.506125271320343, | |
| "eval_runtime": 27.1471, | |
| "eval_samples_per_second": 36.836, | |
| "eval_steps_per_second": 2.321, | |
| "step": 6516 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.0007852056476365869, | |
| "loss": 0.5657, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0007698588090853285, | |
| "loss": 0.5623, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0007545119705340699, | |
| "loss": 0.5669, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.000739195825659914, | |
| "loss": 0.5793, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0007238489871086557, | |
| "loss": 0.5743, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0007085328422344997, | |
| "loss": 0.5722, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0006931860036832413, | |
| "loss": 0.5749, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0006778391651319828, | |
| "loss": 0.5776, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0006624923265807244, | |
| "loss": 0.5694, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.0006471454880294659, | |
| "loss": 0.5716, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.0006317986494782076, | |
| "loss": 0.5665, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.000616451810926949, | |
| "loss": 0.5671, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0006011049723756906, | |
| "loss": 0.5687, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 3.7119, | |
| "eval_gen_len": 18.222, | |
| "eval_loss": 0.495604932308197, | |
| "eval_runtime": 23.8741, | |
| "eval_samples_per_second": 41.886, | |
| "eval_steps_per_second": 2.639, | |
| "step": 13032 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.0005857888275015347, | |
| "loss": 0.5307, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0005704419889502763, | |
| "loss": 0.5325, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.0005550951503990178, | |
| "loss": 0.5353, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 0.0005397483118477594, | |
| "loss": 0.5289, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.0005244014732965009, | |
| "loss": 0.5321, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.000509085328422345, | |
| "loss": 0.5271, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0004937384898710866, | |
| "loss": 0.5295, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.00047839165131982814, | |
| "loss": 0.524, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0004630448127685697, | |
| "loss": 0.526, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.00044772866789441376, | |
| "loss": 0.5323, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00043238182934315537, | |
| "loss": 0.5236, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 0.00041703499079189687, | |
| "loss": 0.5266, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.0004016881522406384, | |
| "loss": 0.525, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 3.8722, | |
| "eval_gen_len": 18.219, | |
| "eval_loss": 0.4868793785572052, | |
| "eval_runtime": 23.2067, | |
| "eval_samples_per_second": 43.091, | |
| "eval_steps_per_second": 2.715, | |
| "step": 19548 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.00038637200736648254, | |
| "loss": 0.4934, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 0.0003710251688152241, | |
| "loss": 0.4869, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.0003556783302639656, | |
| "loss": 0.4917, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 0.00034033149171270715, | |
| "loss": 0.4841, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 0.00032501534683855126, | |
| "loss": 0.4906, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 0.0003096685082872928, | |
| "loss": 0.4884, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 0.0002943216697360344, | |
| "loss": 0.486, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 0.00027897483118477593, | |
| "loss": 0.4932, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.00026365868631062, | |
| "loss": 0.4951, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 0.0002483118477593616, | |
| "loss": 0.4896, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.00023296500920810313, | |
| "loss": 0.4832, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 0.00021761817065684468, | |
| "loss": 0.4874, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00020230202578268877, | |
| "loss": 0.488, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 3.7886, | |
| "eval_gen_len": 18.23, | |
| "eval_loss": 0.4795362949371338, | |
| "eval_runtime": 23.2422, | |
| "eval_samples_per_second": 43.025, | |
| "eval_steps_per_second": 2.711, | |
| "step": 26064 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 0.00018695518723143033, | |
| "loss": 0.4583, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.00017160834868017188, | |
| "loss": 0.4495, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.00015626151012891346, | |
| "loss": 0.454, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.00014094536525475752, | |
| "loss": 0.4547, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 0.00012559852670349908, | |
| "loss": 0.4588, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 0.00011025168815224063, | |
| "loss": 0.453, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 9.49048496009822e-05, | |
| "loss": 0.4549, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 7.955801104972376e-05, | |
| "loss": 0.4533, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 6.424186617556783e-05, | |
| "loss": 0.4495, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.8895027624309394e-05, | |
| "loss": 0.4555, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 3.3548189073050956e-05, | |
| "loss": 0.4563, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 1.820135052179251e-05, | |
| "loss": 0.4563, | |
| "step": 32000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 32580, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.6969603388276736e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |