| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.99722479185939, | |
| "eval_steps": 30, | |
| "global_step": 810, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 3.157894736842105e-06, | |
| "loss": 11.8786, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 11.730859756469727, | |
| "eval_runtime": 60.9075, | |
| "eval_samples_per_second": 0.542, | |
| "eval_steps_per_second": 0.542, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 6.31578947368421e-06, | |
| "loss": 11.386, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 10.65054702758789, | |
| "eval_runtime": 60.8988, | |
| "eval_samples_per_second": 0.542, | |
| "eval_steps_per_second": 0.542, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.473684210526315e-06, | |
| "loss": 9.8984, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 8.621472358703613, | |
| "eval_runtime": 60.8346, | |
| "eval_samples_per_second": 0.542, | |
| "eval_steps_per_second": 0.542, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.263157894736842e-05, | |
| "loss": 7.9555, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 6.96145486831665, | |
| "eval_runtime": 60.9577, | |
| "eval_samples_per_second": 0.541, | |
| "eval_steps_per_second": 0.541, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.578947368421053e-05, | |
| "loss": 6.6507, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 6.004063606262207, | |
| "eval_runtime": 60.9978, | |
| "eval_samples_per_second": 0.541, | |
| "eval_steps_per_second": 0.541, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.894736842105263e-05, | |
| "loss": 5.8685, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 5.406551361083984, | |
| "eval_runtime": 60.6662, | |
| "eval_samples_per_second": 0.544, | |
| "eval_steps_per_second": 0.544, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.9993250234920638e-05, | |
| "loss": 5.3925, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 5.084054946899414, | |
| "eval_runtime": 61.1004, | |
| "eval_samples_per_second": 0.54, | |
| "eval_steps_per_second": 0.54, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.9957838880989076e-05, | |
| "loss": 5.1386, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 4.918034553527832, | |
| "eval_runtime": 60.9721, | |
| "eval_samples_per_second": 0.541, | |
| "eval_steps_per_second": 0.541, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.989218589765658e-05, | |
| "loss": 5.0205, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 4.811317443847656, | |
| "eval_runtime": 60.7077, | |
| "eval_samples_per_second": 0.544, | |
| "eval_steps_per_second": 0.544, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.979649067087574e-05, | |
| "loss": 4.9088, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 4.734280109405518, | |
| "eval_runtime": 61.3359, | |
| "eval_samples_per_second": 0.538, | |
| "eval_steps_per_second": 0.538, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 1.967104382390511e-05, | |
| "loss": 4.8469, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_loss": 4.680157661437988, | |
| "eval_runtime": 60.9063, | |
| "eval_samples_per_second": 0.542, | |
| "eval_steps_per_second": 0.542, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 1.951622633469592e-05, | |
| "loss": 4.7796, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "eval_loss": 4.634165287017822, | |
| "eval_runtime": 61.2818, | |
| "eval_samples_per_second": 0.538, | |
| "eval_steps_per_second": 0.538, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.933250837887457e-05, | |
| "loss": 4.7591, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_loss": 4.599093437194824, | |
| "eval_runtime": 61.1388, | |
| "eval_samples_per_second": 0.54, | |
| "eval_steps_per_second": 0.54, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 1.9120447901834708e-05, | |
| "loss": 4.7235, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "eval_loss": 4.5706963539123535, | |
| "eval_runtime": 60.7134, | |
| "eval_samples_per_second": 0.544, | |
| "eval_steps_per_second": 0.544, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 1.888068892427538e-05, | |
| "loss": 6.7181, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 5.124680995941162, | |
| "eval_runtime": 55.4445, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.595, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.8613959586331364e-05, | |
| "loss": 5.0297, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_loss": 4.7116570472717285, | |
| "eval_runtime": 54.0242, | |
| "eval_samples_per_second": 0.611, | |
| "eval_steps_per_second": 0.611, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.8321069936235503e-05, | |
| "loss": 4.7846, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_loss": 4.612066268920898, | |
| "eval_runtime": 55.4504, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.595, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.800290947022884e-05, | |
| "loss": 4.733, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 4.56680965423584, | |
| "eval_runtime": 55.4115, | |
| "eval_samples_per_second": 0.596, | |
| "eval_steps_per_second": 0.596, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 4.6863, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "eval_loss": 4.538435459136963, | |
| "eval_runtime": 55.5112, | |
| "eval_samples_per_second": 0.594, | |
| "eval_steps_per_second": 0.594, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 1.729471487418621e-05, | |
| "loss": 4.6477, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "eval_loss": 4.5178914070129395, | |
| "eval_runtime": 55.4747, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.595, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 1.6906831507862446e-05, | |
| "loss": 4.6356, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_loss": 4.49837589263916, | |
| "eval_runtime": 55.4914, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.595, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1.64979723212536e-05, | |
| "loss": 4.6119, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "eval_loss": 4.483817100524902, | |
| "eval_runtime": 55.3919, | |
| "eval_samples_per_second": 0.596, | |
| "eval_steps_per_second": 0.596, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 1.606937900627157e-05, | |
| "loss": 4.5994, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "eval_loss": 4.471835613250732, | |
| "eval_runtime": 55.4563, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.595, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 1.5622353186727542e-05, | |
| "loss": 4.5703, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "eval_loss": 4.460899353027344, | |
| "eval_runtime": 55.4262, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.595, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 1.5158252465343242e-05, | |
| "loss": 4.5625, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "eval_loss": 4.450038909912109, | |
| "eval_runtime": 55.4182, | |
| "eval_samples_per_second": 0.595, | |
| "eval_steps_per_second": 0.595, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 1.467848630075608e-05, | |
| "loss": 4.5649, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "eval_loss": 4.4421467781066895, | |
| "eval_runtime": 55.3928, | |
| "eval_samples_per_second": 0.596, | |
| "eval_steps_per_second": 0.596, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.4184511727039612e-05, | |
| "loss": 4.559, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 4.432287693023682, | |
| "eval_runtime": 54.0497, | |
| "eval_samples_per_second": 0.611, | |
| "eval_steps_per_second": 0.611, | |
| "step": 810 | |
| } | |
| ], | |
| "logging_steps": 30, | |
| "max_steps": 1900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 30, | |
| "total_flos": 3.3050813792256e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |