| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.1339111328125, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.7371, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.11322021484375, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.6861, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.127685546875, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.6717, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.13330078125, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.6614, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.135009765625, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.6516, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.1427001953125, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.6427, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.1669921875, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.6333, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.189697265625, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.6269, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.2042236328125, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.6195, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.1954345703125, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6129, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.2440185546875, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.608, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.2342529296875, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.6016, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.24462890625, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.5983, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.2171630859375, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.5935, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.2298583984375, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.5902, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.2283935546875, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.588, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.2369384765625, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.5873, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.2303466796875, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.5858, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.233154296875, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.5856, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.2259521484375, | |
| "learning_rate": 0.0, | |
| "loss": 0.5852, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 200, | |
| "total_flos": 4.038507274174464e+16, | |
| "train_loss": 0.6233242058753967, | |
| "train_runtime": 560.8022, | |
| "train_samples_per_second": 11.412, | |
| "train_steps_per_second": 0.357 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 4.038507274174464e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |