| { | |
| "best_metric": 0.49125435948371887, | |
| "best_model_checkpoint": "../../experiments_checkpoints/MAdAiLab/google_t5/t5_base_twitter/checkpoint-250", | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 816, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.5012608766555786, | |
| "learning_rate": 0.0004938725490196079, | |
| "loss": 0.6085, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.0885683298110962, | |
| "learning_rate": 0.0004877450980392157, | |
| "loss": 0.5402, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 3.7890007495880127, | |
| "learning_rate": 0.00048161764705882356, | |
| "loss": 0.5591, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.2080425024032593, | |
| "learning_rate": 0.00047549019607843134, | |
| "loss": 0.5687, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.9302119612693787, | |
| "learning_rate": 0.0004693627450980392, | |
| "loss": 0.4808, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.7444852941176471, | |
| "eval_f1_macro": 0.6739663773514942, | |
| "eval_f1_micro": 0.7444852941176471, | |
| "eval_loss": 0.5169704556465149, | |
| "eval_runtime": 2.6255, | |
| "eval_samples_per_second": 414.395, | |
| "eval_steps_per_second": 12.95, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.9698123931884766, | |
| "learning_rate": 0.0004632352941176471, | |
| "loss": 0.4972, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.0578800439834595, | |
| "learning_rate": 0.0004571078431372549, | |
| "loss": 0.4967, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 2.512136697769165, | |
| "learning_rate": 0.0004509803921568628, | |
| "loss": 0.4987, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 1.629062533378601, | |
| "learning_rate": 0.00044485294117647056, | |
| "loss": 0.4347, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.581597089767456, | |
| "learning_rate": 0.00043872549019607844, | |
| "loss": 0.5169, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.7555147058823529, | |
| "eval_f1_macro": 0.7269323671497585, | |
| "eval_f1_micro": 0.7555147058823529, | |
| "eval_loss": 0.5100224018096924, | |
| "eval_runtime": 2.6367, | |
| "eval_samples_per_second": 412.631, | |
| "eval_steps_per_second": 12.895, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.149016261100769, | |
| "learning_rate": 0.0004325980392156863, | |
| "loss": 0.4898, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.1472299098968506, | |
| "learning_rate": 0.0004264705882352941, | |
| "loss": 0.497, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.7374740839004517, | |
| "learning_rate": 0.000420343137254902, | |
| "loss": 0.4811, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 1.1296411752700806, | |
| "learning_rate": 0.0004142156862745098, | |
| "loss": 0.5066, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.9138337969779968, | |
| "learning_rate": 0.00040808823529411766, | |
| "loss": 0.4548, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_f1_macro": 0.7017083911650301, | |
| "eval_f1_micro": 0.7647058823529411, | |
| "eval_loss": 0.49222531914711, | |
| "eval_runtime": 2.6971, | |
| "eval_samples_per_second": 403.396, | |
| "eval_steps_per_second": 12.606, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 1.1084121465682983, | |
| "learning_rate": 0.0004019607843137255, | |
| "loss": 0.4683, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 2.0122885704040527, | |
| "learning_rate": 0.0003958333333333333, | |
| "loss": 0.5217, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 2.2070980072021484, | |
| "learning_rate": 0.0003897058823529412, | |
| "loss": 0.524, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.27259361743927, | |
| "learning_rate": 0.00038357843137254904, | |
| "loss": 0.4955, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.8282164931297302, | |
| "learning_rate": 0.0003774509803921569, | |
| "loss": 0.498, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.7518382352941176, | |
| "eval_f1_macro": 0.6776331672629108, | |
| "eval_f1_micro": 0.7518382352941176, | |
| "eval_loss": 0.5056816935539246, | |
| "eval_runtime": 2.6992, | |
| "eval_samples_per_second": 403.083, | |
| "eval_steps_per_second": 12.596, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.8028568029403687, | |
| "learning_rate": 0.0003713235294117647, | |
| "loss": 0.4926, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 2.6962406635284424, | |
| "learning_rate": 0.00036519607843137254, | |
| "loss": 0.5134, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.7592838406562805, | |
| "learning_rate": 0.0003590686274509804, | |
| "loss": 0.469, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.637353777885437, | |
| "learning_rate": 0.00035294117647058826, | |
| "loss": 0.4381, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.9144633412361145, | |
| "learning_rate": 0.0003468137254901961, | |
| "loss": 0.4844, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.765625, | |
| "eval_f1_macro": 0.7266124240384777, | |
| "eval_f1_micro": 0.765625, | |
| "eval_loss": 0.49125435948371887, | |
| "eval_runtime": 2.6614, | |
| "eval_samples_per_second": 408.809, | |
| "eval_steps_per_second": 12.775, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.6250616312026978, | |
| "learning_rate": 0.0003406862745098039, | |
| "loss": 0.424, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.6987557411193848, | |
| "learning_rate": 0.00033455882352941176, | |
| "loss": 0.5254, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.9285285472869873, | |
| "learning_rate": 0.0003284313725490196, | |
| "loss": 0.4381, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 1.1627569198608398, | |
| "learning_rate": 0.0003223039215686275, | |
| "loss": 0.3653, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.7716933488845825, | |
| "learning_rate": 0.0003161764705882353, | |
| "loss": 0.3949, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.7481617647058824, | |
| "eval_f1_macro": 0.6885478829212845, | |
| "eval_f1_micro": 0.7481617647058824, | |
| "eval_loss": 0.5400605201721191, | |
| "eval_runtime": 2.7023, | |
| "eval_samples_per_second": 402.615, | |
| "eval_steps_per_second": 12.582, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 1.0258417129516602, | |
| "learning_rate": 0.00031004901960784314, | |
| "loss": 0.3679, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.483994722366333, | |
| "learning_rate": 0.00030392156862745097, | |
| "loss": 0.3685, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 0.6842905282974243, | |
| "learning_rate": 0.0002977941176470588, | |
| "loss": 0.4533, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.1023057699203491, | |
| "learning_rate": 0.0002916666666666667, | |
| "loss": 0.3752, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 1.9032913446426392, | |
| "learning_rate": 0.0002855392156862745, | |
| "loss": 0.4028, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.7481617647058824, | |
| "eval_f1_macro": 0.7209065444359561, | |
| "eval_f1_micro": 0.7481617647058824, | |
| "eval_loss": 0.5463477373123169, | |
| "eval_runtime": 2.705, | |
| "eval_samples_per_second": 402.218, | |
| "eval_steps_per_second": 12.569, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 1.553853988647461, | |
| "learning_rate": 0.00027941176470588236, | |
| "loss": 0.4068, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 1.6380698680877686, | |
| "learning_rate": 0.0002732843137254902, | |
| "loss": 0.3824, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.9034827947616577, | |
| "learning_rate": 0.000267156862745098, | |
| "loss": 0.4318, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 1.1275397539138794, | |
| "learning_rate": 0.0002610294117647059, | |
| "loss": 0.3843, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 1.784464716911316, | |
| "learning_rate": 0.00025490196078431374, | |
| "loss": 0.3778, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.7555147058823529, | |
| "eval_f1_macro": 0.7087419287313013, | |
| "eval_f1_micro": 0.7555147058823529, | |
| "eval_loss": 0.5437767505645752, | |
| "eval_runtime": 2.7056, | |
| "eval_samples_per_second": 402.135, | |
| "eval_steps_per_second": 12.567, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 1.8430761098861694, | |
| "learning_rate": 0.00024877450980392157, | |
| "loss": 0.4424, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 2.5702669620513916, | |
| "learning_rate": 0.0002426470588235294, | |
| "loss": 0.3864, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 1.207284927368164, | |
| "learning_rate": 0.00023651960784313726, | |
| "loss": 0.4018, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 2.1779820919036865, | |
| "learning_rate": 0.0002303921568627451, | |
| "loss": 0.3279, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 1.1213287115097046, | |
| "learning_rate": 0.00022426470588235296, | |
| "loss": 0.4383, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.7380514705882353, | |
| "eval_f1_macro": 0.7095168638163571, | |
| "eval_f1_micro": 0.7380514705882353, | |
| "eval_loss": 0.5411675572395325, | |
| "eval_runtime": 2.7065, | |
| "eval_samples_per_second": 402.001, | |
| "eval_steps_per_second": 12.563, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 1.428109049797058, | |
| "learning_rate": 0.0002181372549019608, | |
| "loss": 0.3877, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 1.0778878927230835, | |
| "learning_rate": 0.00021200980392156862, | |
| "loss": 0.4038, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.8310728073120117, | |
| "learning_rate": 0.00020588235294117645, | |
| "loss": 0.3548, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.4109036922454834, | |
| "learning_rate": 0.00019975490196078434, | |
| "loss": 0.3681, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 1.2154396772384644, | |
| "learning_rate": 0.00019362745098039217, | |
| "loss": 0.3984, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_accuracy": 0.7555147058823529, | |
| "eval_f1_macro": 0.7239316239316239, | |
| "eval_f1_micro": 0.7555147058823529, | |
| "eval_loss": 0.529325008392334, | |
| "eval_runtime": 2.7052, | |
| "eval_samples_per_second": 402.188, | |
| "eval_steps_per_second": 12.568, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.6760998964309692, | |
| "learning_rate": 0.0001875, | |
| "loss": 0.4037, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 1.0472333431243896, | |
| "learning_rate": 0.00018137254901960784, | |
| "loss": 0.4015, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 1.1614878177642822, | |
| "learning_rate": 0.00017524509803921567, | |
| "loss": 0.4191, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 1.12571382522583, | |
| "learning_rate": 0.00016911764705882356, | |
| "loss": 0.3741, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 0.934890866279602, | |
| "learning_rate": 0.0001629901960784314, | |
| "loss": 0.3122, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.7564338235294118, | |
| "eval_f1_macro": 0.7211537996566648, | |
| "eval_f1_micro": 0.7564338235294118, | |
| "eval_loss": 0.5271954536437988, | |
| "eval_runtime": 2.6516, | |
| "eval_samples_per_second": 410.319, | |
| "eval_steps_per_second": 12.822, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 1.299796462059021, | |
| "learning_rate": 0.00015686274509803922, | |
| "loss": 0.3406, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 0.8316746950149536, | |
| "learning_rate": 0.00015073529411764705, | |
| "loss": 0.2448, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 1.9803074598312378, | |
| "learning_rate": 0.0001446078431372549, | |
| "loss": 0.2767, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 1.0188449621200562, | |
| "learning_rate": 0.00013848039215686274, | |
| "loss": 0.2699, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.744954526424408, | |
| "learning_rate": 0.0001323529411764706, | |
| "loss": 0.2764, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.7463235294117647, | |
| "eval_f1_macro": 0.704812834224599, | |
| "eval_f1_micro": 0.7463235294117647, | |
| "eval_loss": 0.5961496233940125, | |
| "eval_runtime": 2.6549, | |
| "eval_samples_per_second": 409.802, | |
| "eval_steps_per_second": 12.806, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 1.4756693840026855, | |
| "learning_rate": 0.00012622549019607844, | |
| "loss": 0.2649, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 1.1534539461135864, | |
| "learning_rate": 0.00012009803921568628, | |
| "loss": 0.2314, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 1.0833938121795654, | |
| "learning_rate": 0.00011397058823529411, | |
| "loss": 0.2612, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.9921144843101501, | |
| "learning_rate": 0.00010784313725490197, | |
| "loss": 0.2903, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 1.7594339847564697, | |
| "learning_rate": 0.0001017156862745098, | |
| "loss": 0.236, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.7454044117647058, | |
| "eval_f1_macro": 0.6995610690257088, | |
| "eval_f1_micro": 0.7454044117647058, | |
| "eval_loss": 0.6630216240882874, | |
| "eval_runtime": 2.7035, | |
| "eval_samples_per_second": 402.447, | |
| "eval_steps_per_second": 12.576, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 1.5883187055587769, | |
| "learning_rate": 9.558823529411764e-05, | |
| "loss": 0.3054, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 1.9597811698913574, | |
| "learning_rate": 8.94607843137255e-05, | |
| "loss": 0.2728, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.9544721245765686, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.269, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 2.1644415855407715, | |
| "learning_rate": 7.720588235294118e-05, | |
| "loss": 0.2378, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 1.3767701387405396, | |
| "learning_rate": 7.107843137254902e-05, | |
| "loss": 0.1996, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.7481617647058824, | |
| "eval_f1_macro": 0.6967447916666667, | |
| "eval_f1_micro": 0.7481617647058824, | |
| "eval_loss": 0.7069829702377319, | |
| "eval_runtime": 2.6525, | |
| "eval_samples_per_second": 410.183, | |
| "eval_steps_per_second": 12.818, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 2.532257556915283, | |
| "learning_rate": 6.495098039215687e-05, | |
| "loss": 0.2777, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 1.5882879495620728, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.2157, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 1.5534424781799316, | |
| "learning_rate": 5.2696078431372553e-05, | |
| "loss": 0.223, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 1.6244162321090698, | |
| "learning_rate": 4.656862745098039e-05, | |
| "loss": 0.2647, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 1.8495365381240845, | |
| "learning_rate": 4.044117647058824e-05, | |
| "loss": 0.2245, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_accuracy": 0.7454044117647058, | |
| "eval_f1_macro": 0.7015650761743022, | |
| "eval_f1_micro": 0.7454044117647058, | |
| "eval_loss": 0.6734161972999573, | |
| "eval_runtime": 2.6532, | |
| "eval_samples_per_second": 410.075, | |
| "eval_steps_per_second": 12.815, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 1.7572370767593384, | |
| "learning_rate": 3.4313725490196084e-05, | |
| "loss": 0.2363, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 3.104624032974243, | |
| "learning_rate": 2.8186274509803924e-05, | |
| "loss": 0.2159, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 2.204385757446289, | |
| "learning_rate": 2.2058823529411766e-05, | |
| "loss": 0.2228, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 0.7112906575202942, | |
| "learning_rate": 1.593137254901961e-05, | |
| "loss": 0.3132, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 1.3646917343139648, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.2903, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_accuracy": 0.7454044117647058, | |
| "eval_f1_macro": 0.6953530052170152, | |
| "eval_f1_micro": 0.7454044117647058, | |
| "eval_loss": 0.6759808659553528, | |
| "eval_runtime": 2.704, | |
| "eval_samples_per_second": 402.361, | |
| "eval_steps_per_second": 12.574, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 1.2018649578094482, | |
| "learning_rate": 3.6764705882352942e-06, | |
| "loss": 0.2706, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 816, | |
| "total_flos": 3987151653961728.0, | |
| "train_loss": 0.3840603100902894, | |
| "train_runtime": 309.7161, | |
| "train_samples_per_second": 84.271, | |
| "train_steps_per_second": 2.635 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 816, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "total_flos": 3987151653961728.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |