| { |
| "best_global_step": 1955, |
| "best_metric": 7.245829777055189, |
| "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": 24.5, |
| "learning_rate": 9.468731595555683e-05, |
| "loss": 10.4606, |
| "step": 391 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_bleu": 1.7721901631132853, |
| "eval_chrf_pp": 10.721070418371525, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 8.3325, |
| "eval_loss": 4.808479309082031, |
| "eval_runtime": 142.2411, |
| "eval_samples_per_second": 14.061, |
| "eval_steps_per_second": 0.443, |
| "step": 391 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 34.75, |
| "learning_rate": 7.115889418285841e-05, |
| "loss": 5.2159, |
| "step": 782 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_bleu": 4.99344400635358, |
| "eval_chrf_pp": 19.429245875766302, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 12.65, |
| "eval_loss": 4.364200592041016, |
| "eval_runtime": 168.8693, |
| "eval_samples_per_second": 11.843, |
| "eval_steps_per_second": 0.373, |
| "step": 782 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 29.25, |
| "learning_rate": 3.851697720368138e-05, |
| "loss": 4.94, |
| "step": 1173 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_bleu": 6.355236806889961, |
| "eval_chrf_pp": 21.924082601267383, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 11.9125, |
| "eval_loss": 4.265341281890869, |
| "eval_runtime": 164.8318, |
| "eval_samples_per_second": 12.134, |
| "eval_steps_per_second": 0.382, |
| "step": 1173 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 18.125, |
| "learning_rate": 1.0820993235262333e-05, |
| "loss": 4.8809, |
| "step": 1564 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_bleu": 6.987146506440621, |
| "eval_chrf_pp": 22.124878850005334, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 11.2745, |
| "eval_loss": 4.254239082336426, |
| "eval_runtime": 147.7782, |
| "eval_samples_per_second": 13.534, |
| "eval_steps_per_second": 0.426, |
| "step": 1564 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 34.75, |
| "learning_rate": 7.311750030436493e-11, |
| "loss": 4.8732, |
| "step": 1955 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_bleu": 7.245829777055189, |
| "eval_chrf_pp": 22.430202666368746, |
| "eval_empty_rate": 0.0, |
| "eval_gen_len": 11.0125, |
| "eval_loss": 4.25389289855957, |
| "eval_runtime": 161.7857, |
| "eval_samples_per_second": 12.362, |
| "eval_steps_per_second": 0.389, |
| "step": 1955 |
| }, |
| { |
| "epoch": 5.0, |
| "step": 1955, |
| "total_flos": 2.914161718512845e+16, |
| "train_loss": 6.074132957360933, |
| "train_runtime": 4000.9447, |
| "train_samples_per_second": 124.97, |
| "train_steps_per_second": 0.489 |
| } |
| ], |
| "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": 2.914161718512845e+16, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
