| { |
| "best_global_step": 1955, |
| "best_metric": 9.755261490997169, |
| "best_model_checkpoint": "/content/mt5_base_baseline_tr_en_100k_5ep/checkpoint-1955", |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 1955, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 3.609375, |
| "learning_rate": 9.468731595555683e-05, |
| "loss": 9.6576, |
| "step": 391 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_bleu": 3.7407176168912133, |
| "eval_chrf_pp": 14.33263610594533, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 4.395, |
| "eval_loss": 4.268214702606201, |
| "eval_runtime": 90.951, |
| "eval_samples_per_second": 21.99, |
| "eval_steps_per_second": 0.693, |
| "step": 391 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 2.359375, |
| "learning_rate": 7.115889418285841e-05, |
| "loss": 4.5467, |
| "step": 782 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_bleu": 8.008237501117689, |
| "eval_chrf_pp": 22.056956001183675, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 6.2845, |
| "eval_loss": 3.9604849815368652, |
| "eval_runtime": 81.5042, |
| "eval_samples_per_second": 24.539, |
| "eval_steps_per_second": 0.773, |
| "step": 782 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 2.015625, |
| "learning_rate": 3.851697720368138e-05, |
| "loss": 4.3588, |
| "step": 1173 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_bleu": 9.557890582850064, |
| "eval_chrf_pp": 24.373378123346228, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 7.1455, |
| "eval_loss": 3.9073450565338135, |
| "eval_runtime": 79.4762, |
| "eval_samples_per_second": 25.165, |
| "eval_steps_per_second": 0.793, |
| "step": 1173 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.03125, |
| "learning_rate": 1.0820993235262333e-05, |
| "loss": 4.327, |
| "step": 1564 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_bleu": 9.670726359576152, |
| "eval_chrf_pp": 24.688349222485826, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 7.2725, |
| "eval_loss": 3.9009361267089844, |
| "eval_runtime": 80.8154, |
| "eval_samples_per_second": 24.748, |
| "eval_steps_per_second": 0.78, |
| "step": 1564 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.0, |
| "learning_rate": 7.311750030436493e-11, |
| "loss": 4.3209, |
| "step": 1955 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_bleu": 9.755261490997169, |
| "eval_chrf_pp": 24.817022146989494, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 7.2195, |
| "eval_loss": 3.8945579528808594, |
| "eval_runtime": 82.5345, |
| "eval_samples_per_second": 24.232, |
| "eval_steps_per_second": 0.763, |
| "step": 1955 |
| }, |
| { |
| "epoch": 5.0, |
| "step": 1955, |
| "total_flos": 2.51499926126592e+16, |
| "train_loss": 5.442207168618126, |
| "train_runtime": 3581.1772, |
| "train_samples_per_second": 139.619, |
| "train_steps_per_second": 0.546 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 1955, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.51499926126592e+16, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
