| { | |
| "best_metric": 0.4980033338069916, | |
| "best_model_checkpoint": "../../experiments_checkpoints/MAdAiLab/google/flan_t5_base_twitter/checkpoint-50", | |
| "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": 1.5092765092849731, | |
| "learning_rate": 0.0004938725490196079, | |
| "loss": 0.5853, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 2.050435781478882, | |
| "learning_rate": 0.0004877450980392157, | |
| "loss": 0.5303, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 2.564333200454712, | |
| "learning_rate": 0.00048161764705882356, | |
| "loss": 0.5528, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.0580081939697266, | |
| "learning_rate": 0.00047549019607843134, | |
| "loss": 0.5186, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.274806261062622, | |
| "learning_rate": 0.0004693627450980392, | |
| "loss": 0.4683, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.765625, | |
| "eval_f1_macro": 0.7274910494600209, | |
| "eval_f1_micro": 0.765625, | |
| "eval_loss": 0.4980033338069916, | |
| "eval_runtime": 3.0123, | |
| "eval_samples_per_second": 361.184, | |
| "eval_steps_per_second": 11.287, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.2350311279296875, | |
| "learning_rate": 0.0004632352941176471, | |
| "loss": 0.4985, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.238688349723816, | |
| "learning_rate": 0.0004571078431372549, | |
| "loss": 0.4864, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 2.0093843936920166, | |
| "learning_rate": 0.0004509803921568628, | |
| "loss": 0.4836, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 1.2151559591293335, | |
| "learning_rate": 0.00044485294117647056, | |
| "loss": 0.458, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.6234703063964844, | |
| "learning_rate": 0.00043872549019607844, | |
| "loss": 0.5241, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.7582720588235294, | |
| "eval_f1_macro": 0.7287382342541164, | |
| "eval_f1_micro": 0.7582720588235294, | |
| "eval_loss": 0.5018450617790222, | |
| "eval_runtime": 3.0237, | |
| "eval_samples_per_second": 359.823, | |
| "eval_steps_per_second": 11.244, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.0740729570388794, | |
| "learning_rate": 0.0004325980392156863, | |
| "loss": 0.5105, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.214935541152954, | |
| "learning_rate": 0.0004264705882352941, | |
| "loss": 0.4946, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.2495511770248413, | |
| "learning_rate": 0.000420343137254902, | |
| "loss": 0.462, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 1.5260485410690308, | |
| "learning_rate": 0.0004142156862745098, | |
| "loss": 0.5227, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.3698198795318604, | |
| "learning_rate": 0.00040808823529411766, | |
| "loss": 0.4367, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.7564338235294118, | |
| "eval_f1_macro": 0.6810277208948305, | |
| "eval_f1_micro": 0.7564338235294118, | |
| "eval_loss": 0.5173827409744263, | |
| "eval_runtime": 3.033, | |
| "eval_samples_per_second": 358.725, | |
| "eval_steps_per_second": 11.21, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 1.0891311168670654, | |
| "learning_rate": 0.0004019607843137255, | |
| "loss": 0.4669, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.4805020093917847, | |
| "learning_rate": 0.0003958333333333333, | |
| "loss": 0.5305, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.3479547500610352, | |
| "learning_rate": 0.0003897058823529412, | |
| "loss": 0.5154, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.1003795862197876, | |
| "learning_rate": 0.00038357843137254904, | |
| "loss": 0.4962, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.0502173900604248, | |
| "learning_rate": 0.0003774509803921569, | |
| "loss": 0.4868, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.7490808823529411, | |
| "eval_f1_macro": 0.695327189081901, | |
| "eval_f1_micro": 0.7490808823529411, | |
| "eval_loss": 0.5048360228538513, | |
| "eval_runtime": 3.0382, | |
| "eval_samples_per_second": 358.107, | |
| "eval_steps_per_second": 11.191, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.9742279052734375, | |
| "learning_rate": 0.0003713235294117647, | |
| "loss": 0.4764, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 2.2106645107269287, | |
| "learning_rate": 0.00036519607843137254, | |
| "loss": 0.5063, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.7889574766159058, | |
| "learning_rate": 0.0003590686274509804, | |
| "loss": 0.4689, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.676556408405304, | |
| "learning_rate": 0.00035294117647058826, | |
| "loss": 0.4483, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.182094693183899, | |
| "learning_rate": 0.0003468137254901961, | |
| "loss": 0.4824, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.7545955882352942, | |
| "eval_f1_macro": 0.7132815160263528, | |
| "eval_f1_micro": 0.7545955882352942, | |
| "eval_loss": 0.5126115083694458, | |
| "eval_runtime": 3.0384, | |
| "eval_samples_per_second": 358.088, | |
| "eval_steps_per_second": 11.19, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.6538687944412231, | |
| "learning_rate": 0.0003406862745098039, | |
| "loss": 0.4162, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.750243604183197, | |
| "learning_rate": 0.00033455882352941176, | |
| "loss": 0.5175, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.4954441785812378, | |
| "learning_rate": 0.0003284313725490196, | |
| "loss": 0.441, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 1.2219609022140503, | |
| "learning_rate": 0.0003223039215686275, | |
| "loss": 0.354, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.8662141561508179, | |
| "learning_rate": 0.0003161764705882353, | |
| "loss": 0.4008, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.75, | |
| "eval_f1_macro": 0.7067094804860161, | |
| "eval_f1_micro": 0.75, | |
| "eval_loss": 0.5365704894065857, | |
| "eval_runtime": 3.0443, | |
| "eval_samples_per_second": 357.39, | |
| "eval_steps_per_second": 11.168, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 0.8905866146087646, | |
| "learning_rate": 0.00031004901960784314, | |
| "loss": 0.3744, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.7313600778579712, | |
| "learning_rate": 0.00030392156862745097, | |
| "loss": 0.3296, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 0.8115060925483704, | |
| "learning_rate": 0.0002977941176470588, | |
| "loss": 0.4732, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.17222261428833, | |
| "learning_rate": 0.0002916666666666667, | |
| "loss": 0.3701, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 0.9669880867004395, | |
| "learning_rate": 0.0002855392156862745, | |
| "loss": 0.3972, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.7279411764705882, | |
| "eval_f1_macro": 0.6977233135099418, | |
| "eval_f1_micro": 0.7279411764705882, | |
| "eval_loss": 0.5493308901786804, | |
| "eval_runtime": 3.0428, | |
| "eval_samples_per_second": 357.566, | |
| "eval_steps_per_second": 11.174, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 1.2415670156478882, | |
| "learning_rate": 0.00027941176470588236, | |
| "loss": 0.3653, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 2.286729574203491, | |
| "learning_rate": 0.0002732843137254902, | |
| "loss": 0.3603, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.094009280204773, | |
| "learning_rate": 0.000267156862745098, | |
| "loss": 0.4615, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 0.9113813042640686, | |
| "learning_rate": 0.0002610294117647059, | |
| "loss": 0.3903, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 1.8263704776763916, | |
| "learning_rate": 0.00025490196078431374, | |
| "loss": 0.3889, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.7637867647058824, | |
| "eval_f1_macro": 0.7163329508713944, | |
| "eval_f1_micro": 0.7637867647058824, | |
| "eval_loss": 0.5429306030273438, | |
| "eval_runtime": 3.0433, | |
| "eval_samples_per_second": 357.509, | |
| "eval_steps_per_second": 11.172, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 1.1813833713531494, | |
| "learning_rate": 0.00024877450980392157, | |
| "loss": 0.4215, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 1.8237628936767578, | |
| "learning_rate": 0.0002426470588235294, | |
| "loss": 0.3536, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 1.337441086769104, | |
| "learning_rate": 0.00023651960784313726, | |
| "loss": 0.3962, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 1.8698618412017822, | |
| "learning_rate": 0.0002303921568627451, | |
| "loss": 0.3342, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 1.4117869138717651, | |
| "learning_rate": 0.00022426470588235296, | |
| "loss": 0.4151, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_accuracy": 0.7490808823529411, | |
| "eval_f1_macro": 0.7160661999822198, | |
| "eval_f1_micro": 0.7490808823529411, | |
| "eval_loss": 0.5364983677864075, | |
| "eval_runtime": 3.0413, | |
| "eval_samples_per_second": 357.743, | |
| "eval_steps_per_second": 11.179, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 1.9636564254760742, | |
| "learning_rate": 0.0002181372549019608, | |
| "loss": 0.373, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 1.8975811004638672, | |
| "learning_rate": 0.00021200980392156862, | |
| "loss": 0.3889, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 1.7080246210098267, | |
| "learning_rate": 0.00020588235294117645, | |
| "loss": 0.3495, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.1755714416503906, | |
| "learning_rate": 0.00019975490196078434, | |
| "loss": 0.3661, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 1.639256238937378, | |
| "learning_rate": 0.00019362745098039217, | |
| "loss": 0.3842, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_accuracy": 0.75, | |
| "eval_f1_macro": 0.7184907842868666, | |
| "eval_f1_micro": 0.75, | |
| "eval_loss": 0.5349798798561096, | |
| "eval_runtime": 3.0415, | |
| "eval_samples_per_second": 357.716, | |
| "eval_steps_per_second": 11.179, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.3755793571472168, | |
| "learning_rate": 0.0001875, | |
| "loss": 0.4437, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 1.4269018173217773, | |
| "learning_rate": 0.00018137254901960784, | |
| "loss": 0.4263, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 1.2029131650924683, | |
| "learning_rate": 0.00017524509803921567, | |
| "loss": 0.4145, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 1.048883080482483, | |
| "learning_rate": 0.00016911764705882356, | |
| "loss": 0.3648, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 0.9863032102584839, | |
| "learning_rate": 0.0001629901960784314, | |
| "loss": 0.3276, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.7564338235294118, | |
| "eval_f1_macro": 0.7223909884910498, | |
| "eval_f1_micro": 0.7564338235294118, | |
| "eval_loss": 0.5359387397766113, | |
| "eval_runtime": 3.039, | |
| "eval_samples_per_second": 358.017, | |
| "eval_steps_per_second": 11.188, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 1.1191493272781372, | |
| "learning_rate": 0.00015686274509803922, | |
| "loss": 0.3376, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 1.004733681678772, | |
| "learning_rate": 0.00015073529411764705, | |
| "loss": 0.26, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 2.2562036514282227, | |
| "learning_rate": 0.0001446078431372549, | |
| "loss": 0.2736, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 1.4733184576034546, | |
| "learning_rate": 0.00013848039215686274, | |
| "loss": 0.2466, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 1.0705807209014893, | |
| "learning_rate": 0.0001323529411764706, | |
| "loss": 0.3135, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.7564338235294118, | |
| "eval_f1_macro": 0.7120873963341872, | |
| "eval_f1_micro": 0.7564338235294118, | |
| "eval_loss": 0.6310513615608215, | |
| "eval_runtime": 3.0409, | |
| "eval_samples_per_second": 357.789, | |
| "eval_steps_per_second": 11.181, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 1.9386248588562012, | |
| "learning_rate": 0.00012622549019607844, | |
| "loss": 0.2654, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 1.9174495935440063, | |
| "learning_rate": 0.00012009803921568628, | |
| "loss": 0.2348, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 4.020888328552246, | |
| "learning_rate": 0.00011397058823529411, | |
| "loss": 0.2742, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 2.1899900436401367, | |
| "learning_rate": 0.00010784313725490197, | |
| "loss": 0.31, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 1.9852439165115356, | |
| "learning_rate": 0.0001017156862745098, | |
| "loss": 0.2369, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.7490808823529411, | |
| "eval_f1_macro": 0.7048950673768484, | |
| "eval_f1_micro": 0.7490808823529411, | |
| "eval_loss": 0.6068652868270874, | |
| "eval_runtime": 3.0376, | |
| "eval_samples_per_second": 358.18, | |
| "eval_steps_per_second": 11.193, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 2.2759952545166016, | |
| "learning_rate": 9.558823529411764e-05, | |
| "loss": 0.2773, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 1.9017343521118164, | |
| "learning_rate": 8.94607843137255e-05, | |
| "loss": 0.2468, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.6225106716156006, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.268, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 1.8427222967147827, | |
| "learning_rate": 7.720588235294118e-05, | |
| "loss": 0.2501, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 1.5418189764022827, | |
| "learning_rate": 7.107843137254902e-05, | |
| "loss": 0.208, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_accuracy": 0.7536764705882353, | |
| "eval_f1_macro": 0.7090758160492004, | |
| "eval_f1_micro": 0.7536764705882353, | |
| "eval_loss": 0.6517214179039001, | |
| "eval_runtime": 3.0424, | |
| "eval_samples_per_second": 357.618, | |
| "eval_steps_per_second": 11.176, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 2.459115743637085, | |
| "learning_rate": 6.495098039215687e-05, | |
| "loss": 0.2884, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 1.4037036895751953, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.2304, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 1.80465567111969, | |
| "learning_rate": 5.2696078431372553e-05, | |
| "loss": 0.2145, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 4.031944751739502, | |
| "learning_rate": 4.656862745098039e-05, | |
| "loss": 0.2753, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 2.259483814239502, | |
| "learning_rate": 4.044117647058824e-05, | |
| "loss": 0.2667, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_accuracy": 0.7545955882352942, | |
| "eval_f1_macro": 0.7151215187713478, | |
| "eval_f1_micro": 0.7545955882352942, | |
| "eval_loss": 0.6563965678215027, | |
| "eval_runtime": 3.0386, | |
| "eval_samples_per_second": 358.058, | |
| "eval_steps_per_second": 11.189, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 1.8323283195495605, | |
| "learning_rate": 3.4313725490196084e-05, | |
| "loss": 0.2648, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 2.718439817428589, | |
| "learning_rate": 2.8186274509803924e-05, | |
| "loss": 0.2613, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 1.7432453632354736, | |
| "learning_rate": 2.2058823529411766e-05, | |
| "loss": 0.2136, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 1.44099760055542, | |
| "learning_rate": 1.593137254901961e-05, | |
| "loss": 0.278, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 1.8403719663619995, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.3232, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_accuracy": 0.7628676470588235, | |
| "eval_f1_macro": 0.7184852374839538, | |
| "eval_f1_micro": 0.7628676470588235, | |
| "eval_loss": 0.651547908782959, | |
| "eval_runtime": 3.0405, | |
| "eval_samples_per_second": 357.841, | |
| "eval_steps_per_second": 11.183, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 3.050947666168213, | |
| "learning_rate": 3.6764705882352942e-06, | |
| "loss": 0.2703, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 816, | |
| "total_flos": 3987151653961728.0, | |
| "train_loss": 0.38329169475564767, | |
| "train_runtime": 337.4804, | |
| "train_samples_per_second": 77.338, | |
| "train_steps_per_second": 2.418 | |
| } | |
| ], | |
| "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 | |
| } | |