| { | |
| "best_metric": 13.7182, | |
| "best_model_checkpoint": "saved/tobyoki-pairwise/bart-base-japanese/BaseModel/checkpoint-8664", | |
| "epoch": 5.0, | |
| "global_step": 21660, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.8845798707294554e-05, | |
| "loss": 2.1628, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.7691597414589107e-05, | |
| "loss": 1.3515, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.653739612188366e-05, | |
| "loss": 0.9577, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.538319482917821e-05, | |
| "loss": 0.7042, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.422899353647276e-05, | |
| "loss": 0.5935, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.3074792243767315e-05, | |
| "loss": 0.4658, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.192059095106187e-05, | |
| "loss": 0.3854, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.076638965835642e-05, | |
| "loss": 0.2994, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 55.063291139240505, | |
| "eval_loss": 2.7883288860321045, | |
| "eval_rouge1": 11.1611, | |
| "eval_rouge2": 1.7768, | |
| "eval_rougeL": 7.5158, | |
| "eval_rougeLsum": 9.6222, | |
| "eval_runtime": 352.9978, | |
| "eval_samples_per_second": 1.79, | |
| "eval_steps_per_second": 1.79, | |
| "step": 4332 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 3.961218836565097e-05, | |
| "loss": 0.2806, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.845798707294552e-05, | |
| "loss": 0.263, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.7303785780240075e-05, | |
| "loss": 0.2187, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 3.614958448753463e-05, | |
| "loss": 0.1761, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.499538319482918e-05, | |
| "loss": 0.1664, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 3.384118190212373e-05, | |
| "loss": 0.1896, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 3.2686980609418284e-05, | |
| "loss": 0.1627, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.1532779316712836e-05, | |
| "loss": 0.1146, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.0378578024007388e-05, | |
| "loss": 0.1513, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 57.35284810126582, | |
| "eval_loss": 3.128607988357544, | |
| "eval_rouge1": 13.7182, | |
| "eval_rouge2": 2.311, | |
| "eval_rougeL": 9.1726, | |
| "eval_rougeLsum": 11.5058, | |
| "eval_runtime": 358.8793, | |
| "eval_samples_per_second": 1.761, | |
| "eval_steps_per_second": 1.761, | |
| "step": 8664 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.922437673130194e-05, | |
| "loss": 0.125, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.8070175438596492e-05, | |
| "loss": 0.1028, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 2.6915974145891044e-05, | |
| "loss": 0.0934, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 2.5761772853185596e-05, | |
| "loss": 0.0993, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 2.460757156048015e-05, | |
| "loss": 0.074, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.3453370267774704e-05, | |
| "loss": 0.0857, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 2.2299168975069256e-05, | |
| "loss": 0.0823, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 2.1144967682363804e-05, | |
| "loss": 0.0778, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 48.70886075949367, | |
| "eval_loss": 3.3238439559936523, | |
| "eval_rouge1": 12.1173, | |
| "eval_rouge2": 1.88, | |
| "eval_rougeL": 8.1156, | |
| "eval_rougeLsum": 10.1187, | |
| "eval_runtime": 315.5777, | |
| "eval_samples_per_second": 2.003, | |
| "eval_steps_per_second": 2.003, | |
| "step": 12996 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.9990766389658356e-05, | |
| "loss": 0.078, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 1.883656509695291e-05, | |
| "loss": 0.0546, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 1.768236380424746e-05, | |
| "loss": 0.062, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 1.6528162511542013e-05, | |
| "loss": 0.0656, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1.5373961218836565e-05, | |
| "loss": 0.0555, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 1.4219759926131118e-05, | |
| "loss": 0.0437, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.306555863342567e-05, | |
| "loss": 0.05, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 1.1911357340720223e-05, | |
| "loss": 0.0577, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.0757156048014775e-05, | |
| "loss": 0.056, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 50.7373417721519, | |
| "eval_loss": 3.4031858444213867, | |
| "eval_rouge1": 11.9555, | |
| "eval_rouge2": 2.0536, | |
| "eval_rougeL": 8.2185, | |
| "eval_rougeLsum": 10.0656, | |
| "eval_runtime": 343.5063, | |
| "eval_samples_per_second": 1.84, | |
| "eval_steps_per_second": 1.84, | |
| "step": 17328 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 9.602954755309327e-06, | |
| "loss": 0.0416, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 8.448753462603879e-06, | |
| "loss": 0.0393, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 7.29455216989843e-06, | |
| "loss": 0.0368, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 6.140350877192982e-06, | |
| "loss": 0.0433, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 4.986149584487535e-06, | |
| "loss": 0.0421, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 3.831948291782087e-06, | |
| "loss": 0.038, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 2.6777469990766392e-06, | |
| "loss": 0.0285, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 1.5235457063711913e-06, | |
| "loss": 0.0343, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 3.693444136657433e-07, | |
| "loss": 0.0364, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 50.44462025316456, | |
| "eval_loss": 3.525162696838379, | |
| "eval_rouge1": 11.814, | |
| "eval_rouge2": 1.7965, | |
| "eval_rougeL": 8.0177, | |
| "eval_rougeLsum": 9.7342, | |
| "eval_runtime": 338.0819, | |
| "eval_samples_per_second": 1.869, | |
| "eval_steps_per_second": 1.869, | |
| "step": 21660 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 21660, | |
| "total_flos": 3568971203850240.0, | |
| "train_loss": 0.23670565184904047, | |
| "train_runtime": 3830.6771, | |
| "train_samples_per_second": 5.654, | |
| "train_steps_per_second": 5.654 | |
| } | |
| ], | |
| "max_steps": 21660, | |
| "num_train_epochs": 5, | |
| "total_flos": 3568971203850240.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |