| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.934131736526946, | |
| "eval_steps": 500, | |
| "global_step": 123, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11976047904191617, | |
| "grad_norm": 3.5740466117858887, | |
| "learning_rate": 4.979641338636935e-05, | |
| "loss": 2.7298, | |
| "num_input_tokens_seen": 7080, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23952095808383234, | |
| "grad_norm": 3.4822659492492676, | |
| "learning_rate": 4.918896934621734e-05, | |
| "loss": 1.963, | |
| "num_input_tokens_seen": 14016, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3592814371257485, | |
| "grad_norm": 3.181272268295288, | |
| "learning_rate": 4.8187561277552374e-05, | |
| "loss": 1.3739, | |
| "num_input_tokens_seen": 21024, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47904191616766467, | |
| "grad_norm": 2.162691593170166, | |
| "learning_rate": 4.680849904257938e-05, | |
| "loss": 0.9519, | |
| "num_input_tokens_seen": 27864, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5988023952095808, | |
| "grad_norm": 2.1968019008636475, | |
| "learning_rate": 4.507424333013069e-05, | |
| "loss": 0.7551, | |
| "num_input_tokens_seen": 34848, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.718562874251497, | |
| "grad_norm": 1.3013643026351929, | |
| "learning_rate": 4.301303984001967e-05, | |
| "loss": 0.6738, | |
| "num_input_tokens_seen": 41496, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8383233532934131, | |
| "grad_norm": 1.2600340843200684, | |
| "learning_rate": 4.0658459247330766e-05, | |
| "loss": 0.6185, | |
| "num_input_tokens_seen": 48360, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.9580838323353293, | |
| "grad_norm": 1.2103824615478516, | |
| "learning_rate": 3.8048850439214844e-05, | |
| "loss": 0.587, | |
| "num_input_tokens_seen": 55512, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0718562874251496, | |
| "grad_norm": 1.5621482133865356, | |
| "learning_rate": 3.5226715929283506e-05, | |
| "loss": 0.4555, | |
| "num_input_tokens_seen": 61968, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.1916167664670658, | |
| "grad_norm": 1.1218465566635132, | |
| "learning_rate": 3.223801962218372e-05, | |
| "loss": 0.5047, | |
| "num_input_tokens_seen": 68544, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.311377245508982, | |
| "grad_norm": 0.9951040744781494, | |
| "learning_rate": 2.9131438202742124e-05, | |
| "loss": 0.453, | |
| "num_input_tokens_seen": 75456, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.4311377245508983, | |
| "grad_norm": 1.395470380783081, | |
| "learning_rate": 2.595756834225089e-05, | |
| "loss": 0.5011, | |
| "num_input_tokens_seen": 82152, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5508982035928143, | |
| "grad_norm": 1.1427335739135742, | |
| "learning_rate": 2.2768102634070147e-05, | |
| "loss": 0.4825, | |
| "num_input_tokens_seen": 89352, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.6706586826347305, | |
| "grad_norm": 1.1576189994812012, | |
| "learning_rate": 1.961498768002547e-05, | |
| "loss": 0.5319, | |
| "num_input_tokens_seen": 96576, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.7904191616766467, | |
| "grad_norm": 1.177241325378418, | |
| "learning_rate": 1.6549578039787436e-05, | |
| "loss": 0.4653, | |
| "num_input_tokens_seen": 103224, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.910179640718563, | |
| "grad_norm": 1.015809178352356, | |
| "learning_rate": 1.3621799822799788e-05, | |
| "loss": 0.4396, | |
| "num_input_tokens_seen": 110376, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0239520958083834, | |
| "grad_norm": 1.1751824617385864, | |
| "learning_rate": 1.0879337545275165e-05, | |
| "loss": 0.3965, | |
| "num_input_tokens_seen": 116656, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.143712574850299, | |
| "grad_norm": 1.0979323387145996, | |
| "learning_rate": 8.36685749586087e-06, | |
| "loss": 0.4697, | |
| "num_input_tokens_seen": 123904, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.2634730538922154, | |
| "grad_norm": 1.0818545818328857, | |
| "learning_rate": 6.125280258962873e-06, | |
| "loss": 0.4242, | |
| "num_input_tokens_seen": 130552, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.3832335329341316, | |
| "grad_norm": 1.2338745594024658, | |
| "learning_rate": 4.19111424408932e-06, | |
| "loss": 0.3814, | |
| "num_input_tokens_seen": 137200, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.502994011976048, | |
| "grad_norm": 1.1963940858840942, | |
| "learning_rate": 2.595861075973613e-06, | |
| "loss": 0.3991, | |
| "num_input_tokens_seen": 143872, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.622754491017964, | |
| "grad_norm": 1.2328342199325562, | |
| "learning_rate": 1.365502529846166e-06, | |
| "loss": 0.4522, | |
| "num_input_tokens_seen": 150832, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.7425149700598803, | |
| "grad_norm": 1.3839339017868042, | |
| "learning_rate": 5.20077368103597e-07, | |
| "loss": 0.3812, | |
| "num_input_tokens_seen": 157672, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.8622754491017965, | |
| "grad_norm": 1.190239667892456, | |
| "learning_rate": 7.335497040648898e-08, | |
| "loss": 0.4344, | |
| "num_input_tokens_seen": 164776, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.934131736526946, | |
| "num_input_tokens_seen": 169048, | |
| "step": 123, | |
| "total_flos": 992767048187904.0, | |
| "train_loss": 0.696133964430026, | |
| "train_runtime": 82.5141, | |
| "train_samples_per_second": 36.357, | |
| "train_steps_per_second": 1.491 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 123, | |
| "num_input_tokens_seen": 169048, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 992767048187904.0, | |
| "train_batch_size": 3, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |