| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.896, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 3.8122081756591797, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.5688, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 3.92661452293396, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.5951, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 3.3270058631896973, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5533, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 1.8356763124465942, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.5153, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.3816158771514893, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 0.5059, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 2.0332653522491455, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.5338, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 2.24493408203125, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 0.5204, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 1.8572014570236206, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.4666, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 1.762216329574585, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.4972, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.2714998722076416, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.519, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.064, | |
| "grad_norm": 1.70429265499115, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 0.8144, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.950677752494812, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.4161, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.256, | |
| "grad_norm": 1.0790646076202393, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.447, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3519999999999999, | |
| "grad_norm": 0.8673864006996155, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.4061, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.448, | |
| "grad_norm": 0.8862605094909668, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.4371, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.544, | |
| "grad_norm": 0.7694799900054932, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.4347, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.7149304747581482, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.4261, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.736, | |
| "grad_norm": 0.7565575838088989, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.4078, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8319999999999999, | |
| "grad_norm": 0.7304381728172302, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.3913, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.928, | |
| "grad_norm": 0.6986371874809265, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.4079, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.032, | |
| "grad_norm": 1.2996060848236084, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.6794, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.128, | |
| "grad_norm": 0.632396936416626, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.3857, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.224, | |
| "grad_norm": 0.5801264643669128, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.3417, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.5280677080154419, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.327, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.416, | |
| "grad_norm": 0.6040714383125305, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.368, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.512, | |
| "grad_norm": 0.6001883745193481, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.3748, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.608, | |
| "grad_norm": 0.5838417410850525, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.3959, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.7039999999999997, | |
| "grad_norm": 0.6326034665107727, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.3922, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.6273890733718872, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.3947, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "grad_norm": 0.6564163565635681, | |
| "learning_rate": 0.0, | |
| "loss": 0.3837, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "step": 30, | |
| "total_flos": 6484064952320.0, | |
| "train_loss": 0.46355767250061036, | |
| "train_runtime": 870.5984, | |
| "train_samples_per_second": 3.446, | |
| "train_steps_per_second": 0.034 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 30, | |
| "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": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6484064952320.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |