| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.97196261682243, | |
| "eval_steps": 500, | |
| "global_step": 106, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.018691588785046728, | |
| "grad_norm": 4.40625, | |
| "learning_rate": 0.0, | |
| "loss": 2.1468, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09345794392523364, | |
| "grad_norm": 7.625, | |
| "learning_rate": 0.00036363636363636367, | |
| "loss": 1.7296, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.18691588785046728, | |
| "grad_norm": 12.375, | |
| "learning_rate": 0.0008181818181818183, | |
| "loss": 2.261, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2803738317757009, | |
| "grad_norm": 10.25, | |
| "learning_rate": 0.0009975414512725057, | |
| "loss": 3.0423, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.37383177570093457, | |
| "grad_norm": 6.0625, | |
| "learning_rate": 0.0009826044551386743, | |
| "loss": 3.0233, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4672897196261682, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 0.0009545032675245813, | |
| "loss": 2.8968, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5607476635514018, | |
| "grad_norm": 4.4375, | |
| "learning_rate": 0.00091400441557401, | |
| "loss": 3.0136, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6542056074766355, | |
| "grad_norm": 3.5625, | |
| "learning_rate": 0.0008622126023955446, | |
| "loss": 2.9645, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.7476635514018691, | |
| "grad_norm": 3.34375, | |
| "learning_rate": 0.0008005405736415125, | |
| "loss": 2.9556, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8411214953271028, | |
| "grad_norm": 43.0, | |
| "learning_rate": 0.000730670581489344, | |
| "loss": 3.1136, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9345794392523364, | |
| "grad_norm": 13.25, | |
| "learning_rate": 0.0006545084971874737, | |
| "loss": 4.1708, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.325683355331421, | |
| "eval_runtime": 4.4982, | |
| "eval_samples_per_second": 5.335, | |
| "eval_steps_per_second": 5.335, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0186915887850467, | |
| "grad_norm": 5.875, | |
| "learning_rate": 0.000574131823855921, | |
| "loss": 3.1941, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.1121495327102804, | |
| "grad_norm": 4.46875, | |
| "learning_rate": 0.0004917330276168208, | |
| "loss": 2.2848, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.205607476635514, | |
| "grad_norm": 3.03125, | |
| "learning_rate": 0.0004095597328339452, | |
| "loss": 2.1151, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.2990654205607477, | |
| "grad_norm": 2.53125, | |
| "learning_rate": 0.0003298534127791785, | |
| "loss": 2.0086, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.3925233644859814, | |
| "grad_norm": 2.71875, | |
| "learning_rate": 0.0002547882480847461, | |
| "loss": 2.0223, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.485981308411215, | |
| "grad_norm": 7.5625, | |
| "learning_rate": 0.00018641182076323148, | |
| "loss": 1.8961, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.5794392523364484, | |
| "grad_norm": 2.4375, | |
| "learning_rate": 0.0001265892615079232, | |
| "loss": 1.7873, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.6728971962616823, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 7.695237378953224e-05, | |
| "loss": 1.9003, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7663551401869158, | |
| "grad_norm": 1.8046875, | |
| "learning_rate": 3.885512251130763e-05, | |
| "loss": 1.9047, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.8598130841121496, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 1.3336701375997129e-05, | |
| "loss": 1.6384, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.953271028037383, | |
| "grad_norm": 2.671875, | |
| "learning_rate": 1.0931863906127326e-06, | |
| "loss": 1.7658, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.97196261682243, | |
| "eval_loss": 2.7418861389160156, | |
| "eval_runtime": 4.4962, | |
| "eval_samples_per_second": 5.338, | |
| "eval_steps_per_second": 5.338, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.97196261682243, | |
| "step": 106, | |
| "total_flos": 4568119699832832.0, | |
| "train_loss": 2.4587962728626325, | |
| "train_runtime": 293.5611, | |
| "train_samples_per_second": 1.458, | |
| "train_steps_per_second": 0.361 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 106, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 4568119699832832.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |