| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.6879999999999997, | |
| "eval_steps": 30, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 1.7238786220550537, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 10.1751, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 1.528051733970642, | |
| "learning_rate": 2.7142857142857144e-05, | |
| "loss": 10.1038, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 1.8571933507919312, | |
| "learning_rate": 2.5714285714285714e-05, | |
| "loss": 10.1443, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 1.8967256546020508, | |
| "learning_rate": 2.4285714285714288e-05, | |
| "loss": 10.0529, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 2.0033938884735107, | |
| "learning_rate": 2.2857142857142858e-05, | |
| "loss": 9.9815, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 2.3111298084259033, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 10.0506, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 2.3464467525482178, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 10.0345, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "grad_norm": 2.393841028213501, | |
| "learning_rate": 1.8571428571428572e-05, | |
| "loss": 9.9189, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 2.5185725688934326, | |
| "learning_rate": 1.7142857142857142e-05, | |
| "loss": 9.8203, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.6862030029296875, | |
| "learning_rate": 1.5714285714285715e-05, | |
| "loss": 9.7764, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "grad_norm": 2.689467191696167, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 9.689, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 3.0121374130249023, | |
| "learning_rate": 1.2857142857142857e-05, | |
| "loss": 9.6456, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "grad_norm": 3.04107403755188, | |
| "learning_rate": 1.1428571428571429e-05, | |
| "loss": 9.6657, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "grad_norm": 3.180903196334839, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 9.5683, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 3.1552867889404297, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 9.5313, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.048, | |
| "grad_norm": 3.216996908187866, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 9.57, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.176, | |
| "grad_norm": 3.0484824180603027, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 9.4251, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 3.306443214416504, | |
| "learning_rate": 4.2857142857142855e-06, | |
| "loss": 9.4202, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.432, | |
| "grad_norm": 3.5347986221313477, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 9.4979, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 3.746915102005005, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 9.4494, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 3.3043150901794434, | |
| "learning_rate": 0.0, | |
| "loss": 9.3311, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "step": 21, | |
| "total_flos": 284409215227248.0, | |
| "train_loss": 9.754847163245792, | |
| "train_runtime": 2969.0391, | |
| "train_samples_per_second": 0.126, | |
| "train_steps_per_second": 0.007 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 21, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 284409215227248.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |