| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 3e-05, | |
| "loss": 2.9219, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.05291319857312723, | |
| "eval_loss": 2.6484375, | |
| "eval_runtime": 1.1592, | |
| "eval_samples_per_second": 25.016, | |
| "eval_steps_per_second": 1.725, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3e-05, | |
| "loss": 2.6938, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.05291319857312723, | |
| "eval_loss": 2.6484375, | |
| "eval_runtime": 1.6595, | |
| "eval_samples_per_second": 17.475, | |
| "eval_steps_per_second": 1.205, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.99178284305241e-05, | |
| "loss": 2.6365, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.05601796802748051, | |
| "eval_loss": 2.55078125, | |
| "eval_runtime": 1.2012, | |
| "eval_samples_per_second": 24.142, | |
| "eval_steps_per_second": 1.665, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.9672214011007087e-05, | |
| "loss": 2.5088, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.05615008587660193, | |
| "eval_loss": 2.533203125, | |
| "eval_runtime": 0.9032, | |
| "eval_samples_per_second": 32.109, | |
| "eval_steps_per_second": 2.214, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.9265847744427305e-05, | |
| "loss": 2.7307, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_accuracy": 0.05654643942396618, | |
| "eval_loss": 2.517578125, | |
| "eval_runtime": 1.6122, | |
| "eval_samples_per_second": 17.988, | |
| "eval_steps_per_second": 1.241, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.8703181864639013e-05, | |
| "loss": 2.969, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.05714096974501255, | |
| "eval_loss": 2.494140625, | |
| "eval_runtime": 1.5031, | |
| "eval_samples_per_second": 19.294, | |
| "eval_steps_per_second": 1.331, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 2.7990381056766583e-05, | |
| "loss": 2.7283, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.056744616197648305, | |
| "eval_loss": 2.48828125, | |
| "eval_runtime": 0.999, | |
| "eval_samples_per_second": 29.03, | |
| "eval_steps_per_second": 2.002, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.7135254915624213e-05, | |
| "loss": 2.6157, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.05780155899061963, | |
| "eval_loss": 2.4765625, | |
| "eval_runtime": 0.9999, | |
| "eval_samples_per_second": 29.003, | |
| "eval_steps_per_second": 2.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.6147172382160913e-05, | |
| "loss": 2.6406, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.0583300303871053, | |
| "eval_loss": 2.458984375, | |
| "eval_runtime": 1.206, | |
| "eval_samples_per_second": 24.046, | |
| "eval_steps_per_second": 1.658, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.5036959095382875e-05, | |
| "loss": 2.5701, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.05872638393446954, | |
| "eval_loss": 2.4375, | |
| "eval_runtime": 1.0035, | |
| "eval_samples_per_second": 28.898, | |
| "eval_steps_per_second": 1.993, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.3816778784387097e-05, | |
| "loss": 2.2017, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.05866032500990884, | |
| "eval_loss": 2.423828125, | |
| "eval_runtime": 1.1973, | |
| "eval_samples_per_second": 24.222, | |
| "eval_steps_per_second": 1.67, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.25e-05, | |
| "loss": 2.0039, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.05859426608534813, | |
| "eval_loss": 2.421875, | |
| "eval_runtime": 1.009, | |
| "eval_samples_per_second": 28.742, | |
| "eval_steps_per_second": 1.982, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 2.1101049646137008e-05, | |
| "loss": 1.8981, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.058858501783590964, | |
| "eval_loss": 2.416015625, | |
| "eval_runtime": 1.2002, | |
| "eval_samples_per_second": 24.163, | |
| "eval_steps_per_second": 1.666, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.963525491562421e-05, | |
| "loss": 1.7683, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_accuracy": 0.059453032104637336, | |
| "eval_loss": 2.416015625, | |
| "eval_runtime": 1.0984, | |
| "eval_samples_per_second": 26.402, | |
| "eval_steps_per_second": 1.821, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.8118675362266388e-05, | |
| "loss": 1.6746, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.059981503501123, | |
| "eval_loss": 2.412109375, | |
| "eval_runtime": 0.8904, | |
| "eval_samples_per_second": 32.57, | |
| "eval_steps_per_second": 2.246, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.6567926949014805e-05, | |
| "loss": 1.8051, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.06004756242568371, | |
| "eval_loss": 2.41015625, | |
| "eval_runtime": 1.1118, | |
| "eval_samples_per_second": 26.085, | |
| "eval_steps_per_second": 1.799, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.0457, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_accuracy": 0.06017968027480513, | |
| "eval_loss": 2.404296875, | |
| "eval_runtime": 1.1017, | |
| "eval_samples_per_second": 26.322, | |
| "eval_steps_per_second": 1.815, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.3432073050985201e-05, | |
| "loss": 1.8257, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.060576033822169376, | |
| "eval_loss": 2.400390625, | |
| "eval_runtime": 1.5124, | |
| "eval_samples_per_second": 19.175, | |
| "eval_steps_per_second": 1.322, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.1881324637733613e-05, | |
| "loss": 1.744, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.06070815167129079, | |
| "eval_loss": 2.388671875, | |
| "eval_runtime": 0.9026, | |
| "eval_samples_per_second": 32.129, | |
| "eval_steps_per_second": 2.216, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.1881324637733613e-05, | |
| "loss": 1.8232, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.06070815167129079, | |
| "eval_loss": 2.388671875, | |
| "eval_runtime": 1.3479, | |
| "eval_samples_per_second": 21.515, | |
| "eval_steps_per_second": 1.484, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.036474508437579e-05, | |
| "loss": 1.4741, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.06097238736953362, | |
| "eval_loss": 2.3828125, | |
| "eval_runtime": 1.1016, | |
| "eval_samples_per_second": 26.326, | |
| "eval_steps_per_second": 1.816, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 8.898950353863e-06, | |
| "loss": 1.651, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_accuracy": 0.06084026952041221, | |
| "eval_loss": 2.376953125, | |
| "eval_runtime": 1.0964, | |
| "eval_samples_per_second": 26.45, | |
| "eval_steps_per_second": 1.824, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 7.500000000000004e-06, | |
| "loss": 1.3732, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "eval_accuracy": 0.06097238736953362, | |
| "eval_loss": 2.373046875, | |
| "eval_runtime": 1.1007, | |
| "eval_samples_per_second": 26.348, | |
| "eval_steps_per_second": 1.817, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 6.1832212156129045e-06, | |
| "loss": 1.3151, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_accuracy": 0.061038446294094335, | |
| "eval_loss": 2.373046875, | |
| "eval_runtime": 1.0081, | |
| "eval_samples_per_second": 28.767, | |
| "eval_steps_per_second": 1.984, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 4.963040904617131e-06, | |
| "loss": 1.5302, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_accuracy": 0.061038446294094335, | |
| "eval_loss": 2.373046875, | |
| "eval_runtime": 1.2116, | |
| "eval_samples_per_second": 23.936, | |
| "eval_steps_per_second": 1.651, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 3.852827617839085e-06, | |
| "loss": 1.2539, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_accuracy": 0.06117056414321575, | |
| "eval_loss": 2.375, | |
| "eval_runtime": 1.4027, | |
| "eval_samples_per_second": 20.674, | |
| "eval_steps_per_second": 1.426, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 2.86474508437579e-06, | |
| "loss": 1.6211, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_accuracy": 0.061236623067776455, | |
| "eval_loss": 2.376953125, | |
| "eval_runtime": 0.9053, | |
| "eval_samples_per_second": 32.035, | |
| "eval_steps_per_second": 2.209, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 2.0096189432334194e-06, | |
| "loss": 1.6047, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_accuracy": 0.06130268199233716, | |
| "eval_loss": 2.376953125, | |
| "eval_runtime": 1.0962, | |
| "eval_samples_per_second": 26.456, | |
| "eval_steps_per_second": 1.825, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.2968181353609854e-06, | |
| "loss": 1.1953, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.06143479984145858, | |
| "eval_loss": 2.37890625, | |
| "eval_runtime": 1.52, | |
| "eval_samples_per_second": 19.079, | |
| "eval_steps_per_second": 1.316, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 7.341522555726971e-07, | |
| "loss": 1.1621, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.06143479984145858, | |
| "eval_loss": 2.37890625, | |
| "eval_runtime": 1.3108, | |
| "eval_samples_per_second": 22.124, | |
| "eval_steps_per_second": 1.526, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 30, | |
| "total_flos": 1466265894912.0, | |
| "train_loss": 1.999542236328125, | |
| "train_runtime": 244.9371, | |
| "train_samples_per_second": 1.911, | |
| "train_steps_per_second": 0.122 | |
| } | |
| ], | |
| "max_steps": 30, | |
| "num_train_epochs": 3, | |
| "total_flos": 1466265894912.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |