| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 5600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9125000238418579, | |
| "eval_loss": 0.2230108231306076, | |
| "eval_runtime": 5.588, | |
| "eval_samples_per_second": 200.429, | |
| "eval_steps_per_second": 25.054, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.131486410954287e-05, | |
| "loss": 0.3039, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9071428775787354, | |
| "eval_loss": 0.26908189058303833, | |
| "eval_runtime": 5.6261, | |
| "eval_samples_per_second": 199.073, | |
| "eval_steps_per_second": 24.884, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9160714149475098, | |
| "eval_loss": 0.28345081210136414, | |
| "eval_runtime": 5.3742, | |
| "eval_samples_per_second": 208.405, | |
| "eval_steps_per_second": 26.051, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 1.9225171549783764e-05, | |
| "loss": 0.104, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9241071343421936, | |
| "eval_loss": 0.39826202392578125, | |
| "eval_runtime": 5.6203, | |
| "eval_samples_per_second": 199.277, | |
| "eval_steps_per_second": 24.91, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9178571701049805, | |
| "eval_loss": 0.463891863822937, | |
| "eval_runtime": 5.6398, | |
| "eval_samples_per_second": 198.588, | |
| "eval_steps_per_second": 24.823, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 1.713547899002466e-05, | |
| "loss": 0.0361, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9205357432365417, | |
| "eval_loss": 0.46613049507141113, | |
| "eval_runtime": 5.5849, | |
| "eval_samples_per_second": 200.541, | |
| "eval_steps_per_second": 25.068, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9258928298950195, | |
| "eval_loss": 0.4626975357532501, | |
| "eval_runtime": 5.6519, | |
| "eval_samples_per_second": 198.163, | |
| "eval_steps_per_second": 24.77, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 1.5045786430265557e-05, | |
| "loss": 0.0228, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9258928298950195, | |
| "eval_loss": 0.4725589156150818, | |
| "eval_runtime": 5.5505, | |
| "eval_samples_per_second": 201.783, | |
| "eval_steps_per_second": 25.223, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 1.2956093870506452e-05, | |
| "loss": 0.0158, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.925000011920929, | |
| "eval_loss": 0.5453544855117798, | |
| "eval_runtime": 5.5094, | |
| "eval_samples_per_second": 203.288, | |
| "eval_steps_per_second": 25.411, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9160714149475098, | |
| "eval_loss": 0.5454272627830505, | |
| "eval_runtime": 5.4705, | |
| "eval_samples_per_second": 204.734, | |
| "eval_steps_per_second": 25.592, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 1.0866401310747346e-05, | |
| "loss": 0.0146, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9196428656578064, | |
| "eval_loss": 0.5470275282859802, | |
| "eval_runtime": 5.5797, | |
| "eval_samples_per_second": 200.729, | |
| "eval_steps_per_second": 25.091, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9178571701049805, | |
| "eval_loss": 0.6207221746444702, | |
| "eval_runtime": 5.4687, | |
| "eval_samples_per_second": 204.802, | |
| "eval_steps_per_second": 25.6, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 8.776708750988241e-06, | |
| "loss": 0.0106, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9142857193946838, | |
| "eval_loss": 0.6711627840995789, | |
| "eval_runtime": 5.5881, | |
| "eval_samples_per_second": 200.427, | |
| "eval_steps_per_second": 25.053, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.918749988079071, | |
| "eval_loss": 0.6122187972068787, | |
| "eval_runtime": 5.7115, | |
| "eval_samples_per_second": 196.094, | |
| "eval_steps_per_second": 24.512, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 6.687016191229136e-06, | |
| "loss": 0.0101, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9196428656578064, | |
| "eval_loss": 0.6081933975219727, | |
| "eval_runtime": 5.6756, | |
| "eval_samples_per_second": 197.336, | |
| "eval_steps_per_second": 24.667, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9214285612106323, | |
| "eval_loss": 0.6368692517280579, | |
| "eval_runtime": 5.4708, | |
| "eval_samples_per_second": 204.723, | |
| "eval_steps_per_second": 25.59, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 4.59732363147003e-06, | |
| "loss": 0.0108, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9205357432365417, | |
| "eval_loss": 0.6168231964111328, | |
| "eval_runtime": 5.8239, | |
| "eval_samples_per_second": 192.312, | |
| "eval_steps_per_second": 24.039, | |
| "step": 4760 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 2.507631071710926e-06, | |
| "loss": 0.0069, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9196428656578064, | |
| "eval_loss": 0.6445040106773376, | |
| "eval_runtime": 5.6834, | |
| "eval_samples_per_second": 197.066, | |
| "eval_steps_per_second": 24.633, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9178571701049805, | |
| "eval_loss": 0.6451646089553833, | |
| "eval_runtime": 5.6811, | |
| "eval_samples_per_second": 197.145, | |
| "eval_steps_per_second": 24.643, | |
| "step": 5320 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 4.17938511951821e-07, | |
| "loss": 0.0097, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9205357432365417, | |
| "eval_loss": 0.6383888721466064, | |
| "eval_runtime": 5.7353, | |
| "eval_samples_per_second": 195.283, | |
| "eval_steps_per_second": 24.41, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 5600, | |
| "total_flos": 5893687640064000.0, | |
| "train_loss": 0.04879493878356048, | |
| "train_runtime": 2058.2783, | |
| "train_samples_per_second": 43.532, | |
| "train_steps_per_second": 2.721 | |
| } | |
| ], | |
| "max_steps": 5600, | |
| "num_train_epochs": 20, | |
| "total_flos": 5893687640064000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |