| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 15.142274856567383, | |
| "learning_rate": 1.7000000000000002e-06, | |
| "loss": 4.5661, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.56865119934082, | |
| "learning_rate": 3.7e-06, | |
| "loss": 4.056, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.08292586621653063, | |
| "eval_chrf": 12.115072823446873, | |
| "eval_loss": 3.7036895751953125, | |
| "eval_runtime": 178.6768, | |
| "eval_samples_per_second": 7.069, | |
| "eval_steps_per_second": 0.056, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 4.7472944259643555, | |
| "learning_rate": 5.7e-06, | |
| "loss": 3.6426, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.151535987854004, | |
| "learning_rate": 7.7e-06, | |
| "loss": 3.4175, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.11532396839386093, | |
| "eval_chrf": 11.26745223357386, | |
| "eval_loss": 3.222604274749756, | |
| "eval_runtime": 179.1154, | |
| "eval_samples_per_second": 7.051, | |
| "eval_steps_per_second": 0.056, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 4.280997276306152, | |
| "learning_rate": 9.7e-06, | |
| "loss": 3.271, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.886098861694336, | |
| "learning_rate": 7.166666666666667e-06, | |
| "loss": 3.1704, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 0.28031201145196877, | |
| "eval_chrf": 12.970874439653402, | |
| "eval_loss": 3.114114999771118, | |
| "eval_runtime": 173.6137, | |
| "eval_samples_per_second": 7.275, | |
| "eval_steps_per_second": 0.058, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 4.001130104064941, | |
| "learning_rate": 3.833333333333334e-06, | |
| "loss": 3.0489, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.64552116394043, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 3.0286, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 0.403603847823992, | |
| "eval_chrf": 13.32793093086563, | |
| "eval_loss": 3.0850913524627686, | |
| "eval_runtime": 173.0697, | |
| "eval_samples_per_second": 7.298, | |
| "eval_steps_per_second": 0.058, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 160, | |
| "total_flos": 2.6372074438656e+18, | |
| "train_loss": 3.52513427734375, | |
| "train_runtime": 5653.8341, | |
| "train_samples_per_second": 7.192, | |
| "train_steps_per_second": 0.028 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 2.6372074438656e+18, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |