| { | |
| "best_metric": 0.4984971284866333, | |
| "best_model_checkpoint": "../../experiments_checkpoints/MAdAiLab/google/flan_t5_small_twitter/checkpoint-200", | |
| "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": 2.320362091064453, | |
| "learning_rate": 0.0004938725490196079, | |
| "loss": 0.6107, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.2164921760559082, | |
| "learning_rate": 0.0004877450980392157, | |
| "loss": 0.585, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.7246904373168945, | |
| "learning_rate": 0.00048161764705882356, | |
| "loss": 0.5643, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.0919380187988281, | |
| "learning_rate": 0.00047549019607843134, | |
| "loss": 0.5213, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.6304311752319336, | |
| "learning_rate": 0.0004693627450980392, | |
| "loss": 0.5055, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.7536764705882353, | |
| "eval_f1_macro": 0.7230132041417308, | |
| "eval_f1_micro": 0.7536764705882353, | |
| "eval_loss": 0.5209915041923523, | |
| "eval_runtime": 1.0764, | |
| "eval_samples_per_second": 1010.761, | |
| "eval_steps_per_second": 31.586, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 3.424161195755005, | |
| "learning_rate": 0.0004632352941176471, | |
| "loss": 0.5007, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.3720046281814575, | |
| "learning_rate": 0.0004571078431372549, | |
| "loss": 0.5132, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 1.943393588066101, | |
| "learning_rate": 0.0004509803921568628, | |
| "loss": 0.4845, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 1.9219595193862915, | |
| "learning_rate": 0.00044485294117647056, | |
| "loss": 0.4548, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 2.8596079349517822, | |
| "learning_rate": 0.00043872549019607844, | |
| "loss": 0.5045, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.7444852941176471, | |
| "eval_f1_macro": 0.6934304201871463, | |
| "eval_f1_micro": 0.7444852941176471, | |
| "eval_loss": 0.501828134059906, | |
| "eval_runtime": 1.1287, | |
| "eval_samples_per_second": 963.908, | |
| "eval_steps_per_second": 30.122, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.045985460281372, | |
| "learning_rate": 0.0004325980392156863, | |
| "loss": 0.4759, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 2.124852180480957, | |
| "learning_rate": 0.0004264705882352941, | |
| "loss": 0.509, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.3049920797348022, | |
| "learning_rate": 0.000420343137254902, | |
| "loss": 0.4681, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 1.5385198593139648, | |
| "learning_rate": 0.0004142156862745098, | |
| "loss": 0.51, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.8772047162055969, | |
| "learning_rate": 0.00040808823529411766, | |
| "loss": 0.4727, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.7242647058823529, | |
| "eval_f1_macro": 0.6002488646773823, | |
| "eval_f1_micro": 0.7242647058823529, | |
| "eval_loss": 0.5356110334396362, | |
| "eval_runtime": 1.1278, | |
| "eval_samples_per_second": 964.717, | |
| "eval_steps_per_second": 30.147, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 1.149582028388977, | |
| "learning_rate": 0.0004019607843137255, | |
| "loss": 0.4921, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 2.1237943172454834, | |
| "learning_rate": 0.0003958333333333333, | |
| "loss": 0.5149, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.629279613494873, | |
| "learning_rate": 0.0003897058823529412, | |
| "loss": 0.5094, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.2666165828704834, | |
| "learning_rate": 0.00038357843137254904, | |
| "loss": 0.4891, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.3141692876815796, | |
| "learning_rate": 0.0003774509803921569, | |
| "loss": 0.4924, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.7472426470588235, | |
| "eval_f1_macro": 0.6953600443112663, | |
| "eval_f1_micro": 0.7472426470588235, | |
| "eval_loss": 0.4984971284866333, | |
| "eval_runtime": 1.078, | |
| "eval_samples_per_second": 1009.235, | |
| "eval_steps_per_second": 31.539, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.6602263450622559, | |
| "learning_rate": 0.0003713235294117647, | |
| "loss": 0.4878, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 3.3005311489105225, | |
| "learning_rate": 0.00036519607843137254, | |
| "loss": 0.5078, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.2845886945724487, | |
| "learning_rate": 0.0003590686274509804, | |
| "loss": 0.4773, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.7171024084091187, | |
| "learning_rate": 0.00035294117647058826, | |
| "loss": 0.4326, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.302473783493042, | |
| "learning_rate": 0.0003468137254901961, | |
| "loss": 0.4847, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.7527573529411765, | |
| "eval_f1_macro": 0.7106608971260056, | |
| "eval_f1_micro": 0.7527573529411765, | |
| "eval_loss": 0.4992017149925232, | |
| "eval_runtime": 1.1293, | |
| "eval_samples_per_second": 963.438, | |
| "eval_steps_per_second": 30.107, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.1889142990112305, | |
| "learning_rate": 0.0003406862745098039, | |
| "loss": 0.4215, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 1.0543372631072998, | |
| "learning_rate": 0.00033455882352941176, | |
| "loss": 0.5115, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 2.2345387935638428, | |
| "learning_rate": 0.0003284313725490196, | |
| "loss": 0.4543, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 1.7095710039138794, | |
| "learning_rate": 0.0003223039215686275, | |
| "loss": 0.3815, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 1.091468334197998, | |
| "learning_rate": 0.0003161764705882353, | |
| "loss": 0.4107, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.75, | |
| "eval_f1_macro": 0.7086223125425839, | |
| "eval_f1_micro": 0.75, | |
| "eval_loss": 0.5264343023300171, | |
| "eval_runtime": 1.1324, | |
| "eval_samples_per_second": 960.833, | |
| "eval_steps_per_second": 30.026, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 1.4771665334701538, | |
| "learning_rate": 0.00031004901960784314, | |
| "loss": 0.4181, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.9134888648986816, | |
| "learning_rate": 0.00030392156862745097, | |
| "loss": 0.378, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 0.9835084676742554, | |
| "learning_rate": 0.0002977941176470588, | |
| "loss": 0.4827, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.08085036277771, | |
| "learning_rate": 0.0002916666666666667, | |
| "loss": 0.3999, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 1.597136378288269, | |
| "learning_rate": 0.0002855392156862745, | |
| "loss": 0.4197, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.7435661764705882, | |
| "eval_f1_macro": 0.7126257342882434, | |
| "eval_f1_micro": 0.7435661764705882, | |
| "eval_loss": 0.5231310725212097, | |
| "eval_runtime": 1.0822, | |
| "eval_samples_per_second": 1005.332, | |
| "eval_steps_per_second": 31.417, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 1.3496707677841187, | |
| "learning_rate": 0.00027941176470588236, | |
| "loss": 0.4006, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 2.2473416328430176, | |
| "learning_rate": 0.0002732843137254902, | |
| "loss": 0.3715, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.1425981521606445, | |
| "learning_rate": 0.000267156862745098, | |
| "loss": 0.502, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 1.080175757408142, | |
| "learning_rate": 0.0002610294117647059, | |
| "loss": 0.3921, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 1.7369410991668701, | |
| "learning_rate": 0.00025490196078431374, | |
| "loss": 0.4002, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.7509191176470589, | |
| "eval_f1_macro": 0.7008802711523263, | |
| "eval_f1_micro": 0.7509191176470589, | |
| "eval_loss": 0.5311830043792725, | |
| "eval_runtime": 1.1348, | |
| "eval_samples_per_second": 958.773, | |
| "eval_steps_per_second": 29.962, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 1.4130823612213135, | |
| "learning_rate": 0.00024877450980392157, | |
| "loss": 0.4369, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 1.7928345203399658, | |
| "learning_rate": 0.0002426470588235294, | |
| "loss": 0.3897, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 1.4081470966339111, | |
| "learning_rate": 0.00023651960784313726, | |
| "loss": 0.3966, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 1.4095189571380615, | |
| "learning_rate": 0.0002303921568627451, | |
| "loss": 0.3822, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 2.2179651260375977, | |
| "learning_rate": 0.00022426470588235296, | |
| "loss": 0.4381, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.7481617647058824, | |
| "eval_f1_macro": 0.6999822874901374, | |
| "eval_f1_micro": 0.7481617647058824, | |
| "eval_loss": 0.5215727686882019, | |
| "eval_runtime": 1.1354, | |
| "eval_samples_per_second": 958.294, | |
| "eval_steps_per_second": 29.947, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 1.261864185333252, | |
| "learning_rate": 0.0002181372549019608, | |
| "loss": 0.3909, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 1.3953834772109985, | |
| "learning_rate": 0.00021200980392156862, | |
| "loss": 0.4049, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 1.1786818504333496, | |
| "learning_rate": 0.00020588235294117645, | |
| "loss": 0.3903, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.1706407070159912, | |
| "learning_rate": 0.00019975490196078434, | |
| "loss": 0.3724, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 1.9597140550613403, | |
| "learning_rate": 0.00019362745098039217, | |
| "loss": 0.4125, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_accuracy": 0.7509191176470589, | |
| "eval_f1_macro": 0.716105501438017, | |
| "eval_f1_micro": 0.7509191176470589, | |
| "eval_loss": 0.5261591672897339, | |
| "eval_runtime": 1.1364, | |
| "eval_samples_per_second": 957.372, | |
| "eval_steps_per_second": 29.918, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.382961630821228, | |
| "learning_rate": 0.0001875, | |
| "loss": 0.4511, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 1.5272341966629028, | |
| "learning_rate": 0.00018137254901960784, | |
| "loss": 0.4201, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 1.3144079446792603, | |
| "learning_rate": 0.00017524509803921567, | |
| "loss": 0.4401, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 1.6088247299194336, | |
| "learning_rate": 0.00016911764705882356, | |
| "loss": 0.4102, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 1.457602858543396, | |
| "learning_rate": 0.0001629901960784314, | |
| "loss": 0.3665, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.7545955882352942, | |
| "eval_f1_macro": 0.7190493000314321, | |
| "eval_f1_micro": 0.7545955882352942, | |
| "eval_loss": 0.5205144286155701, | |
| "eval_runtime": 1.1386, | |
| "eval_samples_per_second": 955.59, | |
| "eval_steps_per_second": 29.862, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 1.6489365100860596, | |
| "learning_rate": 0.00015686274509803922, | |
| "loss": 0.387, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 1.551331877708435, | |
| "learning_rate": 0.00015073529411764705, | |
| "loss": 0.3103, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 2.288947582244873, | |
| "learning_rate": 0.0001446078431372549, | |
| "loss": 0.3649, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 1.4870082139968872, | |
| "learning_rate": 0.00013848039215686274, | |
| "loss": 0.3382, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 1.9287152290344238, | |
| "learning_rate": 0.0001323529411764706, | |
| "loss": 0.3855, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.7536764705882353, | |
| "eval_f1_macro": 0.7160618855460168, | |
| "eval_f1_micro": 0.7536764705882353, | |
| "eval_loss": 0.5672034025192261, | |
| "eval_runtime": 1.1377, | |
| "eval_samples_per_second": 956.304, | |
| "eval_steps_per_second": 29.884, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 1.6499133110046387, | |
| "learning_rate": 0.00012622549019607844, | |
| "loss": 0.3595, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 1.7214213609695435, | |
| "learning_rate": 0.00012009803921568628, | |
| "loss": 0.2884, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 2.0640528202056885, | |
| "learning_rate": 0.00011397058823529411, | |
| "loss": 0.3366, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 2.1024091243743896, | |
| "learning_rate": 0.00010784313725490197, | |
| "loss": 0.3589, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 1.5444495677947998, | |
| "learning_rate": 0.0001017156862745098, | |
| "loss": 0.3125, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.7509191176470589, | |
| "eval_f1_macro": 0.7065648900543882, | |
| "eval_f1_micro": 0.7509191176470589, | |
| "eval_loss": 0.5732391476631165, | |
| "eval_runtime": 1.0847, | |
| "eval_samples_per_second": 1003.018, | |
| "eval_steps_per_second": 31.344, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 2.2737483978271484, | |
| "learning_rate": 9.558823529411764e-05, | |
| "loss": 0.352, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 2.3418612480163574, | |
| "learning_rate": 8.94607843137255e-05, | |
| "loss": 0.3647, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.5297809839248657, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.3322, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 2.407879590988159, | |
| "learning_rate": 7.720588235294118e-05, | |
| "loss": 0.2971, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 6.948758125305176, | |
| "learning_rate": 7.107843137254902e-05, | |
| "loss": 0.2955, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.7555147058823529, | |
| "eval_f1_macro": 0.7122208301016248, | |
| "eval_f1_micro": 0.7555147058823529, | |
| "eval_loss": 0.5928493738174438, | |
| "eval_runtime": 1.1387, | |
| "eval_samples_per_second": 955.456, | |
| "eval_steps_per_second": 29.858, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 1.9339895248413086, | |
| "learning_rate": 6.495098039215687e-05, | |
| "loss": 0.3605, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 1.7344573736190796, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.3154, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 2.3221147060394287, | |
| "learning_rate": 5.2696078431372553e-05, | |
| "loss": 0.3483, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 2.296344518661499, | |
| "learning_rate": 4.656862745098039e-05, | |
| "loss": 0.3768, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 1.7548387050628662, | |
| "learning_rate": 4.044117647058824e-05, | |
| "loss": 0.3556, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_accuracy": 0.7536764705882353, | |
| "eval_f1_macro": 0.7173598644493644, | |
| "eval_f1_micro": 0.7536764705882353, | |
| "eval_loss": 0.5704131126403809, | |
| "eval_runtime": 1.1365, | |
| "eval_samples_per_second": 957.351, | |
| "eval_steps_per_second": 29.917, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 1.6045734882354736, | |
| "learning_rate": 3.4313725490196084e-05, | |
| "loss": 0.3194, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 1.3206483125686646, | |
| "learning_rate": 2.8186274509803924e-05, | |
| "loss": 0.3289, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 1.6699401140213013, | |
| "learning_rate": 2.2058823529411766e-05, | |
| "loss": 0.32, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 1.2462468147277832, | |
| "learning_rate": 1.593137254901961e-05, | |
| "loss": 0.3742, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 1.7137000560760498, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.3578, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_accuracy": 0.7518382352941176, | |
| "eval_f1_macro": 0.7069047400495676, | |
| "eval_f1_micro": 0.7518382352941176, | |
| "eval_loss": 0.5706028938293457, | |
| "eval_runtime": 1.0853, | |
| "eval_samples_per_second": 1002.521, | |
| "eval_steps_per_second": 31.329, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 1.6963423490524292, | |
| "learning_rate": 3.6764705882352942e-06, | |
| "loss": 0.3367, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 816, | |
| "total_flos": 888901800034304.0, | |
| "train_loss": 0.4176732173152998, | |
| "train_runtime": 109.5646, | |
| "train_samples_per_second": 238.216, | |
| "train_steps_per_second": 7.448 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 816, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "total_flos": 888901800034304.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |