| { | |
| "best_metric": 0.47202983498573303, | |
| "best_model_checkpoint": "/home/iais_marenpielka/Bouthaina/res_nw_yem/checkpoint-918", | |
| "epoch": 16.0, | |
| "eval_steps": 500, | |
| "global_step": 2448, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 43.000205993652344, | |
| "learning_rate": 1.53e-05, | |
| "loss": 7.8314, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.010356123410675664, | |
| "eval_loss": 2.359743595123291, | |
| "eval_rouge1": 0.09828786255795381, | |
| "eval_rouge2": 0.0018177296310952823, | |
| "eval_rougeL": 0.09639405964210626, | |
| "eval_runtime": 1.1837, | |
| "eval_samples_per_second": 256.818, | |
| "eval_steps_per_second": 32.102, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3453600406646729, | |
| "learning_rate": 3.06e-05, | |
| "loss": 0.9995, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.11609954154539694, | |
| "eval_loss": 0.5537932515144348, | |
| "eval_rouge1": 0.47363801342590006, | |
| "eval_rouge2": 0.18538347478382655, | |
| "eval_rougeL": 0.4727595492611425, | |
| "eval_runtime": 9.2767, | |
| "eval_samples_per_second": 32.77, | |
| "eval_steps_per_second": 4.096, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2846729755401611, | |
| "learning_rate": 4.5900000000000004e-05, | |
| "loss": 0.4848, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 0.14721751205352876, | |
| "eval_loss": 0.5034472942352295, | |
| "eval_rouge1": 0.500839694175635, | |
| "eval_rouge2": 0.21735831386751384, | |
| "eval_rougeL": 0.5001625449149274, | |
| "eval_runtime": 10.2581, | |
| "eval_samples_per_second": 29.635, | |
| "eval_steps_per_second": 3.704, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.2576143741607666, | |
| "learning_rate": 4.7812500000000003e-05, | |
| "loss": 0.3823, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.1911807573581778, | |
| "eval_loss": 0.4827471375465393, | |
| "eval_rouge1": 0.5330693017126298, | |
| "eval_rouge2": 0.27439720992270494, | |
| "eval_rougeL": 0.531424425624079, | |
| "eval_runtime": 5.4191, | |
| "eval_samples_per_second": 56.098, | |
| "eval_steps_per_second": 7.012, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.7599295377731323, | |
| "learning_rate": 4.482421875e-05, | |
| "loss": 0.293, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 0.20959557277319146, | |
| "eval_loss": 0.4732062816619873, | |
| "eval_rouge1": 0.5619100572084703, | |
| "eval_rouge2": 0.30991934212678557, | |
| "eval_rougeL": 0.5587085347101918, | |
| "eval_runtime": 2.8763, | |
| "eval_samples_per_second": 105.69, | |
| "eval_steps_per_second": 13.211, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.4758834838867188, | |
| "learning_rate": 4.18359375e-05, | |
| "loss": 0.2239, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 0.22970619705356748, | |
| "eval_loss": 0.47202983498573303, | |
| "eval_rouge1": 0.5777164933812552, | |
| "eval_rouge2": 0.33405816844574837, | |
| "eval_rougeL": 0.5758449342460217, | |
| "eval_runtime": 1.7018, | |
| "eval_samples_per_second": 178.629, | |
| "eval_steps_per_second": 22.329, | |
| "step": 918 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.1557574272155762, | |
| "learning_rate": 3.884765625e-05, | |
| "loss": 0.1766, | |
| "step": 1071 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_bleu": 0.23015149161791434, | |
| "eval_loss": 0.47373390197753906, | |
| "eval_rouge1": 0.588487328464713, | |
| "eval_rouge2": 0.3467647604638442, | |
| "eval_rougeL": 0.5872191002290259, | |
| "eval_runtime": 19.3271, | |
| "eval_samples_per_second": 15.729, | |
| "eval_steps_per_second": 1.966, | |
| "step": 1071 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.236377477645874, | |
| "learning_rate": 3.5859375e-05, | |
| "loss": 0.1434, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_bleu": 0.24506232565608202, | |
| "eval_loss": 0.47583380341529846, | |
| "eval_rouge1": 0.5938395263674476, | |
| "eval_rouge2": 0.3647516376509473, | |
| "eval_rougeL": 0.5903133777624114, | |
| "eval_runtime": 12.6597, | |
| "eval_samples_per_second": 24.013, | |
| "eval_steps_per_second": 3.002, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.2275965213775635, | |
| "learning_rate": 3.287109375e-05, | |
| "loss": 0.1202, | |
| "step": 1377 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_bleu": 0.2508515461433435, | |
| "eval_loss": 0.4827924072742462, | |
| "eval_rouge1": 0.6046956358537867, | |
| "eval_rouge2": 0.3682856780453533, | |
| "eval_rougeL": 0.6015063720236549, | |
| "eval_runtime": 30.4628, | |
| "eval_samples_per_second": 9.979, | |
| "eval_steps_per_second": 1.247, | |
| "step": 1377 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.9903411865234375, | |
| "learning_rate": 2.9882812500000002e-05, | |
| "loss": 0.1042, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_bleu": 0.24386094085391116, | |
| "eval_loss": 0.4852388799190521, | |
| "eval_rouge1": 0.5980129828644593, | |
| "eval_rouge2": 0.3703776134768108, | |
| "eval_rougeL": 0.5951273362785527, | |
| "eval_runtime": 23.4553, | |
| "eval_samples_per_second": 12.961, | |
| "eval_steps_per_second": 1.62, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.104041337966919, | |
| "learning_rate": 2.689453125e-05, | |
| "loss": 0.0955, | |
| "step": 1683 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_bleu": 0.25597221159625966, | |
| "eval_loss": 0.4884573221206665, | |
| "eval_rouge1": 0.6115922153776887, | |
| "eval_rouge2": 0.37952408975600893, | |
| "eval_rougeL": 0.6085537919616412, | |
| "eval_runtime": 1.297, | |
| "eval_samples_per_second": 234.387, | |
| "eval_steps_per_second": 29.298, | |
| "step": 1683 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.6220578551292419, | |
| "learning_rate": 2.3906250000000002e-05, | |
| "loss": 0.0875, | |
| "step": 1836 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_bleu": 0.25506926105311084, | |
| "eval_loss": 0.4913596212863922, | |
| "eval_rouge1": 0.6097696683692526, | |
| "eval_rouge2": 0.38019766139150524, | |
| "eval_rougeL": 0.6055686891522601, | |
| "eval_runtime": 2.0886, | |
| "eval_samples_per_second": 145.549, | |
| "eval_steps_per_second": 18.194, | |
| "step": 1836 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.7318098545074463, | |
| "learning_rate": 2.091796875e-05, | |
| "loss": 0.0825, | |
| "step": 1989 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_bleu": 0.26272086017465046, | |
| "eval_loss": 0.4981193542480469, | |
| "eval_rouge1": 0.6144877321856528, | |
| "eval_rouge2": 0.3872195934378641, | |
| "eval_rougeL": 0.6108120100650027, | |
| "eval_runtime": 1.1422, | |
| "eval_samples_per_second": 266.148, | |
| "eval_steps_per_second": 33.268, | |
| "step": 1989 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.794402539730072, | |
| "learning_rate": 1.79296875e-05, | |
| "loss": 0.0787, | |
| "step": 2142 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_bleu": 0.25895474712357536, | |
| "eval_loss": 0.5004830360412598, | |
| "eval_rouge1": 0.6138305476317043, | |
| "eval_rouge2": 0.3796550798534031, | |
| "eval_rougeL": 0.6100541920692137, | |
| "eval_runtime": 1.1033, | |
| "eval_samples_per_second": 275.529, | |
| "eval_steps_per_second": 34.441, | |
| "step": 2142 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.7674385905265808, | |
| "learning_rate": 1.4941406250000001e-05, | |
| "loss": 0.0757, | |
| "step": 2295 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_bleu": 0.2632177488955002, | |
| "eval_loss": 0.5019999146461487, | |
| "eval_rouge1": 0.6170590027553351, | |
| "eval_rouge2": 0.3850023586523127, | |
| "eval_rougeL": 0.6140574753716048, | |
| "eval_runtime": 4.8266, | |
| "eval_samples_per_second": 62.985, | |
| "eval_steps_per_second": 7.873, | |
| "step": 2295 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.7136771082878113, | |
| "learning_rate": 1.1953125000000001e-05, | |
| "loss": 0.074, | |
| "step": 2448 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_bleu": 0.25943804422988964, | |
| "eval_loss": 0.5049206018447876, | |
| "eval_rouge1": 0.6161012779292924, | |
| "eval_rouge2": 0.3828890396119353, | |
| "eval_rougeL": 0.6124852816095869, | |
| "eval_runtime": 1.0975, | |
| "eval_samples_per_second": 276.998, | |
| "eval_steps_per_second": 34.625, | |
| "step": 2448 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "step": 2448, | |
| "total_flos": 1275105116160000.0, | |
| "train_loss": 0.024901744976542354, | |
| "train_runtime": 209.6658, | |
| "train_samples_per_second": 116.376, | |
| "train_steps_per_second": 14.595 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3060, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1275105116160000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |