| { | |
| "best_metric": 8.105906632790287, | |
| "best_model_checkpoint": "/opt/dlami/nvme/AAA/checkpoint-1986", | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 1986, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.04436992481350899, | |
| "learning_rate": 0.0006666666666666668, | |
| "loss": 0.6502, | |
| "step": 331 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 4.361617844387214, | |
| "eval_loss": 0.007830865681171417, | |
| "eval_runtime": 121.9513, | |
| "eval_samples_per_second": 10.849, | |
| "eval_steps_per_second": 0.681, | |
| "step": 331 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.038081083446741104, | |
| "learning_rate": 0.0005333333333333334, | |
| "loss": 0.0071, | |
| "step": 662 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 7.843100767141589, | |
| "eval_loss": 0.0018343101255595684, | |
| "eval_runtime": 122.0918, | |
| "eval_samples_per_second": 10.836, | |
| "eval_steps_per_second": 0.68, | |
| "step": 662 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.019110124558210373, | |
| "learning_rate": 0.00045714285714285713, | |
| "loss": 0.0027, | |
| "step": 993 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_bleu": 8.040733224615437, | |
| "eval_loss": 0.0016475560842081904, | |
| "eval_runtime": 122.063, | |
| "eval_samples_per_second": 10.839, | |
| "eval_steps_per_second": 0.68, | |
| "step": 993 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.006580352783203125, | |
| "learning_rate": 0.00034285714285714285, | |
| "loss": 0.0016, | |
| "step": 1324 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_bleu": 8.046469246678804, | |
| "eval_loss": 0.0006783127901144326, | |
| "eval_runtime": 122.0945, | |
| "eval_samples_per_second": 10.836, | |
| "eval_steps_per_second": 0.68, | |
| "step": 1324 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.0007480424828827381, | |
| "learning_rate": 0.00022857142857142857, | |
| "loss": 0.001, | |
| "step": 1655 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_bleu": 8.065670810538343, | |
| "eval_loss": 0.0004619210085365921, | |
| "eval_runtime": 122.0219, | |
| "eval_samples_per_second": 10.842, | |
| "eval_steps_per_second": 0.68, | |
| "step": 1655 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.00835416093468666, | |
| "learning_rate": 0.00011428571428571428, | |
| "loss": 0.0007, | |
| "step": 1986 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_bleu": 8.105906632790287, | |
| "eval_loss": 0.00035273091634735465, | |
| "eval_runtime": 121.9325, | |
| "eval_samples_per_second": 10.85, | |
| "eval_steps_per_second": 0.681, | |
| "step": 1986 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2317, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.933197719371776e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |