| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 5.582018382418061, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.832, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 6.008115759273852, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.8909, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 4.203158365014558, | |
| "learning_rate": 2e-05, | |
| "loss": 0.856, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 5.330103222720995, | |
| "learning_rate": 1.989821441880933e-05, | |
| "loss": 0.8935, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 6.604796866941393, | |
| "learning_rate": 1.9594929736144978e-05, | |
| "loss": 0.8573, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 4.889759493234221, | |
| "learning_rate": 1.9096319953545186e-05, | |
| "loss": 0.8122, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.6673253563493007, | |
| "learning_rate": 1.8412535328311813e-05, | |
| "loss": 0.7676, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.6708268813136478, | |
| "learning_rate": 1.7557495743542586e-05, | |
| "loss": 0.7087, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.8635895305372137, | |
| "learning_rate": 1.6548607339452853e-05, | |
| "loss": 0.695, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.1743860615942052, | |
| "learning_rate": 1.5406408174555978e-05, | |
| "loss": 0.6507, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.9577383518041913, | |
| "learning_rate": 1.4154150130018867e-05, | |
| "loss": 0.626, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.9653843420006609, | |
| "learning_rate": 1.2817325568414299e-05, | |
| "loss": 0.5799, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.7211363515893529, | |
| "learning_rate": 1.1423148382732854e-05, | |
| "loss": 0.5685, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.90772975706778, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5919, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.7164003657107308, | |
| "learning_rate": 8.576851617267151e-06, | |
| "loss": 0.5679, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.2467863416781333, | |
| "learning_rate": 7.182674431585703e-06, | |
| "loss": 0.498, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.9480737486309809, | |
| "learning_rate": 5.845849869981137e-06, | |
| "loss": 0.5647, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.5516119711486823, | |
| "learning_rate": 4.593591825444028e-06, | |
| "loss": 0.5169, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.7064792732116788, | |
| "learning_rate": 3.4513926605471504e-06, | |
| "loss": 0.5086, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5816794153324469, | |
| "learning_rate": 2.4425042564574186e-06, | |
| "loss": 0.4558, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 0.5315955082840325, | |
| "learning_rate": 1.587464671688187e-06, | |
| "loss": 0.4984, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.4271511280028261, | |
| "learning_rate": 9.036800464548157e-07, | |
| "loss": 0.4525, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 0.45389257042160547, | |
| "learning_rate": 4.0507026385502747e-07, | |
| "loss": 0.4713, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.443007797239538, | |
| "learning_rate": 1.0178558119067316e-07, | |
| "loss": 0.4809, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.44135326132925484, | |
| "learning_rate": 0.0, | |
| "loss": 0.4748, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 25, | |
| "total_flos": 3.2478016827293696e+16, | |
| "train_loss": 0.6328057992458344, | |
| "train_runtime": 2119.4087, | |
| "train_samples_per_second": 0.745, | |
| "train_steps_per_second": 0.012 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 25, | |
| "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": 3.2478016827293696e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |