| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9473684210526314, | |
| "eval_steps": 500, | |
| "global_step": 84, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17543859649122806, | |
| "grad_norm": 8.90099811553955, | |
| "learning_rate": 1.996992941167792e-05, | |
| "loss": 1.1137, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 2.5175933837890625, | |
| "learning_rate": 1.9633708786158803e-05, | |
| "loss": 0.6094, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 2.091320276260376, | |
| "learning_rate": 1.8936326403234125e-05, | |
| "loss": 0.5038, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 1.8124364614486694, | |
| "learning_rate": 1.7903926695187595e-05, | |
| "loss": 0.484, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8771929824561403, | |
| "grad_norm": 1.689386248588562, | |
| "learning_rate": 1.657521368569064e-05, | |
| "loss": 0.4102, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 1.687546730041504, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.3906, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2280701754385965, | |
| "grad_norm": 1.8159677982330322, | |
| "learning_rate": 1.3237339420583213e-05, | |
| "loss": 0.3151, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "grad_norm": 1.6340556144714355, | |
| "learning_rate": 1.1353312997501313e-05, | |
| "loss": 0.2973, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 1.5886197090148926, | |
| "learning_rate": 9.418551710895243e-06, | |
| "loss": 0.2915, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "grad_norm": 1.5932583808898926, | |
| "learning_rate": 7.505588559420188e-06, | |
| "loss": 0.2892, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9298245614035088, | |
| "grad_norm": 1.5423833131790161, | |
| "learning_rate": 5.686139343187468e-06, | |
| "loss": 0.2769, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 1.4709863662719727, | |
| "learning_rate": 4.028414082972141e-06, | |
| "loss": 0.2353, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.280701754385965, | |
| "grad_norm": 1.43242347240448, | |
| "learning_rate": 2.594559868909956e-06, | |
| "loss": 0.1994, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.456140350877193, | |
| "grad_norm": 1.5272055864334106, | |
| "learning_rate": 1.4383310046973365e-06, | |
| "loss": 0.1864, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 1.5854390859603882, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.187, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.807017543859649, | |
| "grad_norm": 1.4928103685379028, | |
| "learning_rate": 1.201015052319099e-07, | |
| "loss": 0.1925, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.9473684210526314, | |
| "step": 84, | |
| "total_flos": 3.214315621240013e+16, | |
| "train_loss": 0.3641331955080941, | |
| "train_runtime": 776.361, | |
| "train_samples_per_second": 7.048, | |
| "train_steps_per_second": 0.108 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 84, | |
| "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": 3.214315621240013e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |