| { | |
| "best_metric": 0.4186987578868866, | |
| "best_model_checkpoint": "/tmp/model/checkpoint-114", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 114, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.0986, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.0986, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.0986, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.098, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.0968, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.0957, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 1.0952, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.0934, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 1.0903, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 1.0896, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.0911, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.0736, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 1.1018, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5e-05, | |
| "loss": 1.075, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.9509803921568634e-05, | |
| "loss": 1.0414, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 4.901960784313725e-05, | |
| "loss": 1.0614, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.8529411764705885e-05, | |
| "loss": 1.0775, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.803921568627452e-05, | |
| "loss": 1.0337, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.7549019607843135e-05, | |
| "loss": 0.9952, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 1.0467, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.656862745098039e-05, | |
| "loss": 1.0225, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.607843137254902e-05, | |
| "loss": 1.0948, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.558823529411765e-05, | |
| "loss": 1.0724, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.558823529411765e-05, | |
| "loss": 1.0368, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.5098039215686275e-05, | |
| "loss": 1.046, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.460784313725491e-05, | |
| "loss": 1.0462, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.411764705882353e-05, | |
| "loss": 1.0037, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 4.362745098039216e-05, | |
| "loss": 1.0037, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.313725490196079e-05, | |
| "loss": 1.0308, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.2647058823529415e-05, | |
| "loss": 1.0509, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.215686274509804e-05, | |
| "loss": 1.0226, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.9627, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.11764705882353e-05, | |
| "loss": 0.9468, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.068627450980392e-05, | |
| "loss": 0.9843, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.0196078431372555e-05, | |
| "loss": 0.7904, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.970588235294117e-05, | |
| "loss": 0.9915, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3.9215686274509805e-05, | |
| "loss": 0.9587, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.872549019607844e-05, | |
| "loss": 1.003, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7105263157894737, | |
| "eval_f1_macro": 0.5220918441257424, | |
| "eval_f1_micro": 0.7105263157894737, | |
| "eval_f1_weighted": 0.6349642032068438, | |
| "eval_loss": 0.8949874043464661, | |
| "eval_precision_macro": 0.4696864111498258, | |
| "eval_precision_micro": 0.7105263157894737, | |
| "eval_precision_weighted": 0.5784705666605539, | |
| "eval_recall_macro": 0.5938438438438438, | |
| "eval_recall_micro": 0.7105263157894737, | |
| "eval_recall_weighted": 0.7105263157894737, | |
| "eval_runtime": 2.0351, | |
| "eval_samples_per_second": 37.344, | |
| "eval_steps_per_second": 2.457, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 3.8235294117647055e-05, | |
| "loss": 1.0132, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.8235294117647055e-05, | |
| "loss": 0.8854, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 3.774509803921569e-05, | |
| "loss": 0.9694, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 3.725490196078432e-05, | |
| "loss": 0.8182, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 3.6764705882352945e-05, | |
| "loss": 0.9112, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 3.627450980392157e-05, | |
| "loss": 0.8819, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 3.5784313725490195e-05, | |
| "loss": 1.0679, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.8977, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.480392156862745e-05, | |
| "loss": 0.7788, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.431372549019608e-05, | |
| "loss": 0.9201, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.382352941176471e-05, | |
| "loss": 0.781, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.8634, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3.284313725490196e-05, | |
| "loss": 0.8995, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 3.235294117647059e-05, | |
| "loss": 0.7176, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 3.186274509803922e-05, | |
| "loss": 1.0019, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 3.137254901960784e-05, | |
| "loss": 0.6014, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 3.0882352941176475e-05, | |
| "loss": 0.6722, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 3.0392156862745097e-05, | |
| "loss": 0.8279, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.9901960784313725e-05, | |
| "loss": 0.7848, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 0.8737, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.8921568627450986e-05, | |
| "loss": 0.679, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 2.8431372549019608e-05, | |
| "loss": 0.5924, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.7941176470588236e-05, | |
| "loss": 0.6882, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 2.7450980392156865e-05, | |
| "loss": 0.5811, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.696078431372549e-05, | |
| "loss": 0.6237, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 2.647058823529412e-05, | |
| "loss": 0.7408, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 2.5980392156862747e-05, | |
| "loss": 0.5923, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.5490196078431373e-05, | |
| "loss": 0.53, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.8175, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.4509803921568626e-05, | |
| "loss": 0.816, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 2.401960784313726e-05, | |
| "loss": 0.6064, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 0.7559, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 2.303921568627451e-05, | |
| "loss": 0.6855, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 2.2549019607843138e-05, | |
| "loss": 0.4998, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 2.2058823529411766e-05, | |
| "loss": 0.3965, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 2.1568627450980395e-05, | |
| "loss": 0.2904, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.107843137254902e-05, | |
| "loss": 0.5538, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.058823529411765e-05, | |
| "loss": 0.6493, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8552631578947368, | |
| "eval_f1_macro": 0.766620230632304, | |
| "eval_f1_micro": 0.8552631578947367, | |
| "eval_f1_weighted": 0.8357707159703299, | |
| "eval_loss": 0.508880615234375, | |
| "eval_precision_macro": 0.8952380952380953, | |
| "eval_precision_micro": 0.8552631578947368, | |
| "eval_precision_weighted": 0.9007518796992481, | |
| "eval_recall_macro": 0.7687687687687688, | |
| "eval_recall_micro": 0.8552631578947368, | |
| "eval_recall_weighted": 0.8552631578947368, | |
| "eval_runtime": 2.0308, | |
| "eval_samples_per_second": 37.425, | |
| "eval_steps_per_second": 2.462, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 2.0098039215686277e-05, | |
| "loss": 0.6863, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.9607843137254903e-05, | |
| "loss": 0.6366, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.9117647058823528e-05, | |
| "loss": 0.6654, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1.862745098039216e-05, | |
| "loss": 0.6877, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.8137254901960785e-05, | |
| "loss": 0.4684, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 0.8363, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.715686274509804e-05, | |
| "loss": 0.49, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3663, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1.6176470588235296e-05, | |
| "loss": 0.4103, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.568627450980392e-05, | |
| "loss": 0.6288, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.5196078431372548e-05, | |
| "loss": 0.5966, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1.4705882352941177e-05, | |
| "loss": 0.5132, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.4215686274509804e-05, | |
| "loss": 0.3739, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 1.3725490196078432e-05, | |
| "loss": 0.6379, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 1.323529411764706e-05, | |
| "loss": 0.7557, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 1.2745098039215686e-05, | |
| "loss": 0.3617, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.2254901960784313e-05, | |
| "loss": 0.2977, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 0.9881, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.1274509803921569e-05, | |
| "loss": 0.7618, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 1.0784313725490197e-05, | |
| "loss": 0.3369, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1.0294117647058824e-05, | |
| "loss": 0.4169, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.6706, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 9.31372549019608e-06, | |
| "loss": 0.634, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 8.823529411764707e-06, | |
| "loss": 0.2243, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.8181, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 7.84313725490196e-06, | |
| "loss": 0.5856, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 7.3529411764705884e-06, | |
| "loss": 0.4225, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 6.862745098039216e-06, | |
| "loss": 0.3595, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 6.372549019607843e-06, | |
| "loss": 0.5012, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.7054, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 5.392156862745099e-06, | |
| "loss": 0.539, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.901960784313726e-06, | |
| "loss": 0.4866, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 4.411764705882353e-06, | |
| "loss": 0.4528, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 3.92156862745098e-06, | |
| "loss": 0.4653, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 3.431372549019608e-06, | |
| "loss": 0.5724, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 2.9411764705882355e-06, | |
| "loss": 0.5088, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 2.450980392156863e-06, | |
| "loss": 0.7226, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.96078431372549e-06, | |
| "loss": 0.2725, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.881578947368421, | |
| "eval_f1_macro": 0.8215900897948484, | |
| "eval_f1_micro": 0.881578947368421, | |
| "eval_f1_weighted": 0.8716990992917834, | |
| "eval_loss": 0.4186987578868866, | |
| "eval_precision_macro": 0.9090909090909092, | |
| "eval_precision_micro": 0.881578947368421, | |
| "eval_precision_weighted": 0.9138755980861244, | |
| "eval_recall_macro": 0.8132132132132132, | |
| "eval_recall_micro": 0.881578947368421, | |
| "eval_recall_weighted": 0.881578947368421, | |
| "eval_runtime": 2.034, | |
| "eval_samples_per_second": 37.365, | |
| "eval_steps_per_second": 2.458, | |
| "step": 114 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 114, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 6.994598554602086e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |