| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 45, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 13.329143873173242, | |
| "learning_rate": 4.993910125649561e-06, | |
| "loss": 0.992, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 9.341056087485404, | |
| "learning_rate": 4.849231551964771e-06, | |
| "loss": 0.7071, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 2.4011701108598835, | |
| "learning_rate": 4.415111107797445e-06, | |
| "loss": 0.2824, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.8721878855044509, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.2054, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.5988037325683389, | |
| "learning_rate": 2.9341204441673267e-06, | |
| "loss": 0.1871, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 0.6463197256589266, | |
| "learning_rate": 2.0658795558326745e-06, | |
| "loss": 0.1666, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.4962059147709929, | |
| "learning_rate": 1.2500000000000007e-06, | |
| "loss": 0.153, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 0.5032120456217833, | |
| "learning_rate": 5.848888922025553e-07, | |
| "loss": 0.1415, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 0.4938264292826346, | |
| "learning_rate": 1.507684480352292e-07, | |
| "loss": 0.1332, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.5250301729177533, | |
| "learning_rate": 0.0, | |
| "loss": 0.1333, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 45, | |
| "total_flos": 6969999785984.0, | |
| "train_loss": 0.24072090254889594, | |
| "train_runtime": 954.7262, | |
| "train_samples_per_second": 3.017, | |
| "train_steps_per_second": 0.047 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 45, | |
| "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": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6969999785984.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |