| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 448, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17857142857142858, | |
| "grad_norm": 17.865875244140625, | |
| "learning_rate": 1.9910714285714287e-05, | |
| "loss": 1.0838, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 3.3213462829589844, | |
| "learning_rate": 1.9151785714285714e-05, | |
| "loss": 0.7877, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "grad_norm": 2.8315253257751465, | |
| "learning_rate": 1.8258928571428574e-05, | |
| "loss": 0.6396, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 2.6161699295043945, | |
| "learning_rate": 1.736607142857143e-05, | |
| "loss": 0.5952, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "grad_norm": 2.6655831336975098, | |
| "learning_rate": 1.6473214285714288e-05, | |
| "loss": 0.5743, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0714285714285714, | |
| "grad_norm": 2.081516981124878, | |
| "learning_rate": 1.5580357142857145e-05, | |
| "loss": 0.4802, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.047880172729492, | |
| "learning_rate": 1.4687500000000001e-05, | |
| "loss": 0.3406, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.9429481029510498, | |
| "learning_rate": 1.379464285714286e-05, | |
| "loss": 0.347, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.6071428571428572, | |
| "grad_norm": 2.0169007778167725, | |
| "learning_rate": 1.2901785714285715e-05, | |
| "loss": 0.333, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.7857142857142856, | |
| "grad_norm": 1.9162710905075073, | |
| "learning_rate": 1.2008928571428573e-05, | |
| "loss": 0.3454, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.9642857142857144, | |
| "grad_norm": 1.9847056865692139, | |
| "learning_rate": 1.1116071428571428e-05, | |
| "loss": 0.3392, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 1.6575144529342651, | |
| "learning_rate": 1.0223214285714287e-05, | |
| "loss": 0.1765, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.3214285714285716, | |
| "grad_norm": 1.6254658699035645, | |
| "learning_rate": 9.330357142857143e-06, | |
| "loss": 0.1455, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.5523484945297241, | |
| "learning_rate": 8.4375e-06, | |
| "loss": 0.1407, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.678571428571429, | |
| "grad_norm": 1.5673795938491821, | |
| "learning_rate": 7.544642857142858e-06, | |
| "loss": 0.1397, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 1.509041428565979, | |
| "learning_rate": 6.6517857142857144e-06, | |
| "loss": 0.1371, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.0357142857142856, | |
| "grad_norm": 1.0347487926483154, | |
| "learning_rate": 5.758928571428571e-06, | |
| "loss": 0.1147, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.2142857142857144, | |
| "grad_norm": 1.1131216287612915, | |
| "learning_rate": 4.866071428571429e-06, | |
| "loss": 0.0511, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.392857142857143, | |
| "grad_norm": 1.2071995735168457, | |
| "learning_rate": 3.9732142857142855e-06, | |
| "loss": 0.0523, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 1.0390278100967407, | |
| "learning_rate": 3.080357142857143e-06, | |
| "loss": 0.05, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.982015073299408, | |
| "learning_rate": 2.1875000000000002e-06, | |
| "loss": 0.0473, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.928571428571429, | |
| "grad_norm": 0.9728031158447266, | |
| "learning_rate": 1.2946428571428574e-06, | |
| "loss": 0.0437, | |
| "step": 440 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 448, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 76855084482560.0, | |
| "train_batch_size": 12, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |