| { | |
| "best_metric": 0.0046388679184019566, | |
| "best_model_checkpoint": "./mobilevitv2_Liveness_detection_v1.0/checkpoint-1792", | |
| "epoch": 3.0, | |
| "eval_steps": 128, | |
| "global_step": 1875, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0256, | |
| "grad_norm": 0.686086893081665, | |
| "learning_rate": 4.9573333333333335e-05, | |
| "loss": 0.6647, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0512, | |
| "grad_norm": 0.7888203859329224, | |
| "learning_rate": 4.914666666666667e-05, | |
| "loss": 0.6063, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0768, | |
| "grad_norm": 0.9471352100372314, | |
| "learning_rate": 4.872000000000001e-05, | |
| "loss": 0.5254, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1024, | |
| "grad_norm": 0.9644356966018677, | |
| "learning_rate": 4.8293333333333334e-05, | |
| "loss": 0.437, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 1.174842119216919, | |
| "learning_rate": 4.7866666666666674e-05, | |
| "loss": 0.3153, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1536, | |
| "grad_norm": 1.0370259284973145, | |
| "learning_rate": 4.744e-05, | |
| "loss": 0.2325, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.1792, | |
| "grad_norm": 0.7825124859809875, | |
| "learning_rate": 4.701333333333334e-05, | |
| "loss": 0.1471, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2048, | |
| "grad_norm": 0.6782599687576294, | |
| "learning_rate": 4.6586666666666666e-05, | |
| "loss": 0.1093, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.2048, | |
| "eval_accuracy": 0.9929, | |
| "eval_f1": 0.992897770729339, | |
| "eval_loss": 0.06788957864046097, | |
| "eval_precision": 0.9929058981404609, | |
| "eval_recall": 0.9929, | |
| "eval_runtime": 247.0578, | |
| "eval_samples_per_second": 80.953, | |
| "eval_steps_per_second": 10.119, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.2304, | |
| "grad_norm": 0.5946743488311768, | |
| "learning_rate": 4.6160000000000005e-05, | |
| "loss": 0.0675, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 0.6635921597480774, | |
| "learning_rate": 4.573333333333333e-05, | |
| "loss": 0.0477, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2816, | |
| "grad_norm": 0.22356025874614716, | |
| "learning_rate": 4.530666666666667e-05, | |
| "loss": 0.0434, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.3072, | |
| "grad_norm": 0.7422251105308533, | |
| "learning_rate": 4.488e-05, | |
| "loss": 0.0332, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.3328, | |
| "grad_norm": 0.3897477090358734, | |
| "learning_rate": 4.445333333333334e-05, | |
| "loss": 0.0294, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.3584, | |
| "grad_norm": 0.7231374979019165, | |
| "learning_rate": 4.402666666666666e-05, | |
| "loss": 0.0251, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.6253751516342163, | |
| "learning_rate": 4.36e-05, | |
| "loss": 0.0218, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4096, | |
| "grad_norm": 0.14728830754756927, | |
| "learning_rate": 4.3173333333333336e-05, | |
| "loss": 0.0234, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.4096, | |
| "eval_accuracy": 0.9962, | |
| "eval_f1": 0.99619730556991, | |
| "eval_loss": 0.016960417851805687, | |
| "eval_precision": 0.9962127076373954, | |
| "eval_recall": 0.9962, | |
| "eval_runtime": 219.599, | |
| "eval_samples_per_second": 91.075, | |
| "eval_steps_per_second": 11.384, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.4352, | |
| "grad_norm": 0.45428574085235596, | |
| "learning_rate": 4.274666666666667e-05, | |
| "loss": 0.0174, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.4608, | |
| "grad_norm": 0.11591943353414536, | |
| "learning_rate": 4.232e-05, | |
| "loss": 0.0195, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.4864, | |
| "grad_norm": 0.34196004271507263, | |
| "learning_rate": 4.1893333333333334e-05, | |
| "loss": 0.0239, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 0.18962785601615906, | |
| "learning_rate": 4.146666666666667e-05, | |
| "loss": 0.0203, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.5376, | |
| "grad_norm": 0.05710538849234581, | |
| "learning_rate": 4.104e-05, | |
| "loss": 0.0247, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.5632, | |
| "grad_norm": 1.2746449708938599, | |
| "learning_rate": 4.061333333333334e-05, | |
| "loss": 0.017, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.5888, | |
| "grad_norm": 0.46703022718429565, | |
| "learning_rate": 4.0186666666666666e-05, | |
| "loss": 0.0112, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.6144, | |
| "grad_norm": 2.4929864406585693, | |
| "learning_rate": 3.9760000000000006e-05, | |
| "loss": 0.0186, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.6144, | |
| "eval_accuracy": 0.9973, | |
| "eval_f1": 0.9973011575273466, | |
| "eval_loss": 0.013053582981228828, | |
| "eval_precision": 0.9973004220662013, | |
| "eval_recall": 0.9973, | |
| "eval_runtime": 215.8374, | |
| "eval_samples_per_second": 92.662, | |
| "eval_steps_per_second": 11.583, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.2726776897907257, | |
| "learning_rate": 3.933333333333333e-05, | |
| "loss": 0.009, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6656, | |
| "grad_norm": 0.04591584578156471, | |
| "learning_rate": 3.890666666666667e-05, | |
| "loss": 0.0137, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 0.6912, | |
| "grad_norm": 0.2672736644744873, | |
| "learning_rate": 3.848e-05, | |
| "loss": 0.0156, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 0.7168, | |
| "grad_norm": 0.07182446122169495, | |
| "learning_rate": 3.805333333333334e-05, | |
| "loss": 0.0168, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 0.7424, | |
| "grad_norm": 0.029036881402134895, | |
| "learning_rate": 3.762666666666667e-05, | |
| "loss": 0.0104, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 1.6431952714920044, | |
| "learning_rate": 3.72e-05, | |
| "loss": 0.0145, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7936, | |
| "grad_norm": 0.5779602527618408, | |
| "learning_rate": 3.6773333333333336e-05, | |
| "loss": 0.0099, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 0.8192, | |
| "grad_norm": 0.07454531639814377, | |
| "learning_rate": 3.634666666666667e-05, | |
| "loss": 0.0068, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 0.8192, | |
| "eval_accuracy": 0.99805, | |
| "eval_f1": 0.9980503428173515, | |
| "eval_loss": 0.008904083631932735, | |
| "eval_precision": 0.9980497728736165, | |
| "eval_recall": 0.99805, | |
| "eval_runtime": 216.611, | |
| "eval_samples_per_second": 92.331, | |
| "eval_steps_per_second": 11.541, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 0.8448, | |
| "grad_norm": 0.0550871342420578, | |
| "learning_rate": 3.592e-05, | |
| "loss": 0.0095, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 0.8704, | |
| "grad_norm": 0.13159841299057007, | |
| "learning_rate": 3.5493333333333335e-05, | |
| "loss": 0.0089, | |
| "step": 544 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 0.6349660754203796, | |
| "learning_rate": 3.506666666666667e-05, | |
| "loss": 0.0096, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.9216, | |
| "grad_norm": 0.7711471915245056, | |
| "learning_rate": 3.464e-05, | |
| "loss": 0.015, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 0.9472, | |
| "grad_norm": 0.05490660294890404, | |
| "learning_rate": 3.424e-05, | |
| "loss": 0.0113, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 0.9728, | |
| "grad_norm": 1.9179311990737915, | |
| "learning_rate": 3.381333333333334e-05, | |
| "loss": 0.0084, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 0.9984, | |
| "grad_norm": 0.04221022129058838, | |
| "learning_rate": 3.338666666666666e-05, | |
| "loss": 0.0166, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "grad_norm": 0.21213921904563904, | |
| "learning_rate": 3.296e-05, | |
| "loss": 0.0049, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "eval_accuracy": 0.9985, | |
| "eval_f1": 0.9984999412798902, | |
| "eval_loss": 0.006704141851514578, | |
| "eval_precision": 0.9985001400955447, | |
| "eval_recall": 0.9985, | |
| "eval_runtime": 215.2961, | |
| "eval_samples_per_second": 92.895, | |
| "eval_steps_per_second": 11.612, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.0496, | |
| "grad_norm": 0.24302862584590912, | |
| "learning_rate": 3.253333333333333e-05, | |
| "loss": 0.0055, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 1.0752, | |
| "grad_norm": 0.17449580132961273, | |
| "learning_rate": 3.210666666666667e-05, | |
| "loss": 0.0078, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 1.1008, | |
| "grad_norm": 1.3184378147125244, | |
| "learning_rate": 3.168e-05, | |
| "loss": 0.0073, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 1.1264, | |
| "grad_norm": 0.8744115829467773, | |
| "learning_rate": 3.1253333333333335e-05, | |
| "loss": 0.0074, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 0.07388653606176376, | |
| "learning_rate": 3.082666666666667e-05, | |
| "loss": 0.0039, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.1776, | |
| "grad_norm": 0.035437047481536865, | |
| "learning_rate": 3.04e-05, | |
| "loss": 0.008, | |
| "step": 736 | |
| }, | |
| { | |
| "epoch": 1.2032, | |
| "grad_norm": 0.06002597510814667, | |
| "learning_rate": 2.9973333333333337e-05, | |
| "loss": 0.0033, | |
| "step": 752 | |
| }, | |
| { | |
| "epoch": 1.2288000000000001, | |
| "grad_norm": 0.01166555192321539, | |
| "learning_rate": 2.9546666666666666e-05, | |
| "loss": 0.0113, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 1.2288000000000001, | |
| "eval_accuracy": 0.99835, | |
| "eval_f1": 0.9983500322782247, | |
| "eval_loss": 0.006390445865690708, | |
| "eval_precision": 0.9983499411063025, | |
| "eval_recall": 0.99835, | |
| "eval_runtime": 215.8826, | |
| "eval_samples_per_second": 92.643, | |
| "eval_steps_per_second": 11.58, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 1.2544, | |
| "grad_norm": 2.266014575958252, | |
| "learning_rate": 2.9120000000000002e-05, | |
| "loss": 0.0089, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.0778322219848633, | |
| "learning_rate": 2.8693333333333332e-05, | |
| "loss": 0.0041, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3056, | |
| "grad_norm": 0.18685859441757202, | |
| "learning_rate": 2.8266666666666668e-05, | |
| "loss": 0.0041, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 1.3312, | |
| "grad_norm": 0.09057345986366272, | |
| "learning_rate": 2.7839999999999998e-05, | |
| "loss": 0.0071, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 1.3568, | |
| "grad_norm": 0.28758856654167175, | |
| "learning_rate": 2.7413333333333334e-05, | |
| "loss": 0.0045, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 1.3824, | |
| "grad_norm": 0.9198317527770996, | |
| "learning_rate": 2.6986666666666667e-05, | |
| "loss": 0.0107, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "grad_norm": 0.14636975526809692, | |
| "learning_rate": 2.6560000000000003e-05, | |
| "loss": 0.004, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.4336, | |
| "grad_norm": 0.038640450686216354, | |
| "learning_rate": 2.6133333333333333e-05, | |
| "loss": 0.0061, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 1.4336, | |
| "eval_accuracy": 0.99835, | |
| "eval_f1": 0.9983498384304894, | |
| "eval_loss": 0.0060436660423874855, | |
| "eval_precision": 0.9983504647077259, | |
| "eval_recall": 0.99835, | |
| "eval_runtime": 219.3162, | |
| "eval_samples_per_second": 91.193, | |
| "eval_steps_per_second": 11.399, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 1.4592, | |
| "grad_norm": 2.163743257522583, | |
| "learning_rate": 2.570666666666667e-05, | |
| "loss": 0.0071, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 1.4848, | |
| "grad_norm": 1.2653355598449707, | |
| "learning_rate": 2.5280000000000005e-05, | |
| "loss": 0.0046, | |
| "step": 928 | |
| }, | |
| { | |
| "epoch": 1.5104, | |
| "grad_norm": 0.025220897048711777, | |
| "learning_rate": 2.4853333333333335e-05, | |
| "loss": 0.0039, | |
| "step": 944 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.015397891402244568, | |
| "learning_rate": 2.4426666666666668e-05, | |
| "loss": 0.0065, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.5615999999999999, | |
| "grad_norm": 0.06891310960054398, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.0049, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 1.5872000000000002, | |
| "grad_norm": 0.23304007947444916, | |
| "learning_rate": 2.3573333333333334e-05, | |
| "loss": 0.0125, | |
| "step": 992 | |
| }, | |
| { | |
| "epoch": 1.6128, | |
| "grad_norm": 0.010659678839147091, | |
| "learning_rate": 2.3146666666666666e-05, | |
| "loss": 0.0044, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 1.6383999999999999, | |
| "grad_norm": 0.0605621375143528, | |
| "learning_rate": 2.2720000000000003e-05, | |
| "loss": 0.0025, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 1.6383999999999999, | |
| "eval_accuracy": 0.99835, | |
| "eval_f1": 0.9983497737194017, | |
| "eval_loss": 0.005817316006869078, | |
| "eval_precision": 0.9983507300357946, | |
| "eval_recall": 0.99835, | |
| "eval_runtime": 214.0685, | |
| "eval_samples_per_second": 93.428, | |
| "eval_steps_per_second": 11.679, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "grad_norm": 0.9562625288963318, | |
| "learning_rate": 2.2293333333333336e-05, | |
| "loss": 0.0081, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.6896, | |
| "grad_norm": 1.08597731590271, | |
| "learning_rate": 2.186666666666667e-05, | |
| "loss": 0.0035, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 1.7151999999999998, | |
| "grad_norm": 0.08577126264572144, | |
| "learning_rate": 2.144e-05, | |
| "loss": 0.004, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 1.7408000000000001, | |
| "grad_norm": 0.007748621515929699, | |
| "learning_rate": 2.1013333333333334e-05, | |
| "loss": 0.0055, | |
| "step": 1088 | |
| }, | |
| { | |
| "epoch": 1.7664, | |
| "grad_norm": 0.5899467468261719, | |
| "learning_rate": 2.0586666666666667e-05, | |
| "loss": 0.0059, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "grad_norm": 1.230574369430542, | |
| "learning_rate": 2.016e-05, | |
| "loss": 0.0052, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.8176, | |
| "grad_norm": 0.4939228296279907, | |
| "learning_rate": 1.9733333333333333e-05, | |
| "loss": 0.0083, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 1.8432, | |
| "grad_norm": 0.02068273536860943, | |
| "learning_rate": 1.9306666666666666e-05, | |
| "loss": 0.0019, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 1.8432, | |
| "eval_accuracy": 0.99865, | |
| "eval_f1": 0.9986499735808153, | |
| "eval_loss": 0.0053438348695635796, | |
| "eval_precision": 0.998650058503282, | |
| "eval_recall": 0.99865, | |
| "eval_runtime": 222.9408, | |
| "eval_samples_per_second": 89.71, | |
| "eval_steps_per_second": 11.214, | |
| "step": 1152 | |
| }, | |
| { | |
| "epoch": 1.8688, | |
| "grad_norm": 1.5578925609588623, | |
| "learning_rate": 1.888e-05, | |
| "loss": 0.0063, | |
| "step": 1168 | |
| }, | |
| { | |
| "epoch": 1.8944, | |
| "grad_norm": 0.00804234016686678, | |
| "learning_rate": 1.8453333333333335e-05, | |
| "loss": 0.0046, | |
| "step": 1184 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 2.660892963409424, | |
| "learning_rate": 1.8026666666666668e-05, | |
| "loss": 0.0046, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.9456, | |
| "grad_norm": 0.008745341561734676, | |
| "learning_rate": 1.76e-05, | |
| "loss": 0.0029, | |
| "step": 1216 | |
| }, | |
| { | |
| "epoch": 1.9712, | |
| "grad_norm": 0.015926124528050423, | |
| "learning_rate": 1.7173333333333334e-05, | |
| "loss": 0.0037, | |
| "step": 1232 | |
| }, | |
| { | |
| "epoch": 1.9968, | |
| "grad_norm": 0.009791089221835136, | |
| "learning_rate": 1.674666666666667e-05, | |
| "loss": 0.0067, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 2.0224, | |
| "grad_norm": 0.009614923968911171, | |
| "learning_rate": 1.6320000000000003e-05, | |
| "loss": 0.0022, | |
| "step": 1264 | |
| }, | |
| { | |
| "epoch": 2.048, | |
| "grad_norm": 0.007969129830598831, | |
| "learning_rate": 1.5893333333333336e-05, | |
| "loss": 0.0056, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.048, | |
| "eval_accuracy": 0.9987, | |
| "eval_f1": 0.9986997451712335, | |
| "eval_loss": 0.005075570661574602, | |
| "eval_precision": 0.9987010761843749, | |
| "eval_recall": 0.9987, | |
| "eval_runtime": 220.6361, | |
| "eval_samples_per_second": 90.647, | |
| "eval_steps_per_second": 11.331, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.0736, | |
| "grad_norm": 0.01278685312718153, | |
| "learning_rate": 1.546666666666667e-05, | |
| "loss": 0.0029, | |
| "step": 1296 | |
| }, | |
| { | |
| "epoch": 2.0992, | |
| "grad_norm": 0.013344179838895798, | |
| "learning_rate": 1.5040000000000002e-05, | |
| "loss": 0.0029, | |
| "step": 1312 | |
| }, | |
| { | |
| "epoch": 2.1248, | |
| "grad_norm": 0.021782569587230682, | |
| "learning_rate": 1.4613333333333335e-05, | |
| "loss": 0.0021, | |
| "step": 1328 | |
| }, | |
| { | |
| "epoch": 2.1504, | |
| "grad_norm": 0.036290887743234634, | |
| "learning_rate": 1.4186666666666667e-05, | |
| "loss": 0.0031, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 2.176, | |
| "grad_norm": 0.7045446634292603, | |
| "learning_rate": 1.376e-05, | |
| "loss": 0.003, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.2016, | |
| "grad_norm": 0.012572742067277431, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0089, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 2.2272, | |
| "grad_norm": 0.05552055686712265, | |
| "learning_rate": 1.2906666666666666e-05, | |
| "loss": 0.0011, | |
| "step": 1392 | |
| }, | |
| { | |
| "epoch": 2.2528, | |
| "grad_norm": 0.047569043934345245, | |
| "learning_rate": 1.248e-05, | |
| "loss": 0.0015, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 2.2528, | |
| "eval_accuracy": 0.9987, | |
| "eval_f1": 0.9987, | |
| "eval_loss": 0.004999413620680571, | |
| "eval_precision": 0.9987, | |
| "eval_recall": 0.9987, | |
| "eval_runtime": 216.7321, | |
| "eval_samples_per_second": 92.28, | |
| "eval_steps_per_second": 11.535, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 2.2784, | |
| "grad_norm": 0.391156405210495, | |
| "learning_rate": 1.2053333333333334e-05, | |
| "loss": 0.0038, | |
| "step": 1424 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.01173941045999527, | |
| "learning_rate": 1.1626666666666668e-05, | |
| "loss": 0.0075, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.3296, | |
| "grad_norm": 2.2623393535614014, | |
| "learning_rate": 1.1200000000000001e-05, | |
| "loss": 0.0093, | |
| "step": 1456 | |
| }, | |
| { | |
| "epoch": 2.3552, | |
| "grad_norm": 0.01715581864118576, | |
| "learning_rate": 1.0773333333333334e-05, | |
| "loss": 0.0057, | |
| "step": 1472 | |
| }, | |
| { | |
| "epoch": 2.3808, | |
| "grad_norm": 1.0069034099578857, | |
| "learning_rate": 1.0346666666666667e-05, | |
| "loss": 0.0011, | |
| "step": 1488 | |
| }, | |
| { | |
| "epoch": 2.4064, | |
| "grad_norm": 0.20940986275672913, | |
| "learning_rate": 9.92e-06, | |
| "loss": 0.0022, | |
| "step": 1504 | |
| }, | |
| { | |
| "epoch": 2.432, | |
| "grad_norm": 0.011296062730252743, | |
| "learning_rate": 9.493333333333333e-06, | |
| "loss": 0.0013, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 2.4576000000000002, | |
| "grad_norm": 1.7626785039901733, | |
| "learning_rate": 9.066666666666667e-06, | |
| "loss": 0.0055, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 2.4576000000000002, | |
| "eval_accuracy": 0.99875, | |
| "eval_f1": 0.9987499265863476, | |
| "eval_loss": 0.004937997553497553, | |
| "eval_precision": 0.998750197817996, | |
| "eval_recall": 0.99875, | |
| "eval_runtime": 216.9352, | |
| "eval_samples_per_second": 92.193, | |
| "eval_steps_per_second": 11.524, | |
| "step": 1536 | |
| }, | |
| { | |
| "epoch": 2.4832, | |
| "grad_norm": 0.029261792078614235, | |
| "learning_rate": 8.64e-06, | |
| "loss": 0.0038, | |
| "step": 1552 | |
| }, | |
| { | |
| "epoch": 2.5088, | |
| "grad_norm": 0.04974502697587013, | |
| "learning_rate": 8.213333333333333e-06, | |
| "loss": 0.0051, | |
| "step": 1568 | |
| }, | |
| { | |
| "epoch": 2.5343999999999998, | |
| "grad_norm": 0.03413296118378639, | |
| "learning_rate": 7.786666666666668e-06, | |
| "loss": 0.0013, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.02191867306828499, | |
| "learning_rate": 7.36e-06, | |
| "loss": 0.0011, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.5856, | |
| "grad_norm": 0.00806102529168129, | |
| "learning_rate": 6.933333333333334e-06, | |
| "loss": 0.0032, | |
| "step": 1616 | |
| }, | |
| { | |
| "epoch": 2.6112, | |
| "grad_norm": 0.0360230915248394, | |
| "learning_rate": 6.5066666666666665e-06, | |
| "loss": 0.007, | |
| "step": 1632 | |
| }, | |
| { | |
| "epoch": 2.6368, | |
| "grad_norm": 0.024547284469008446, | |
| "learning_rate": 6.08e-06, | |
| "loss": 0.007, | |
| "step": 1648 | |
| }, | |
| { | |
| "epoch": 2.6624, | |
| "grad_norm": 0.18587003648281097, | |
| "learning_rate": 5.653333333333333e-06, | |
| "loss": 0.0023, | |
| "step": 1664 | |
| }, | |
| { | |
| "epoch": 2.6624, | |
| "eval_accuracy": 0.99885, | |
| "eval_f1": 0.9988497971544271, | |
| "eval_loss": 0.004876961465924978, | |
| "eval_precision": 0.998850864665525, | |
| "eval_recall": 0.99885, | |
| "eval_runtime": 216.0532, | |
| "eval_samples_per_second": 92.57, | |
| "eval_steps_per_second": 11.571, | |
| "step": 1664 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.8803591728210449, | |
| "learning_rate": 5.226666666666667e-06, | |
| "loss": 0.0034, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 2.7136, | |
| "grad_norm": 0.033573221415281296, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 0.0057, | |
| "step": 1696 | |
| }, | |
| { | |
| "epoch": 2.7392, | |
| "grad_norm": 0.1415340006351471, | |
| "learning_rate": 4.3733333333333335e-06, | |
| "loss": 0.0017, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 2.7648, | |
| "grad_norm": 0.5475676655769348, | |
| "learning_rate": 3.9466666666666664e-06, | |
| "loss": 0.0067, | |
| "step": 1728 | |
| }, | |
| { | |
| "epoch": 2.7904, | |
| "grad_norm": 0.062486432492733, | |
| "learning_rate": 3.52e-06, | |
| "loss": 0.0033, | |
| "step": 1744 | |
| }, | |
| { | |
| "epoch": 2.816, | |
| "grad_norm": 0.009262731298804283, | |
| "learning_rate": 3.0933333333333335e-06, | |
| "loss": 0.0014, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 2.8416, | |
| "grad_norm": 0.07086601853370667, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 0.003, | |
| "step": 1776 | |
| }, | |
| { | |
| "epoch": 2.8672, | |
| "grad_norm": 0.04299961403012276, | |
| "learning_rate": 2.24e-06, | |
| "loss": 0.0027, | |
| "step": 1792 | |
| }, | |
| { | |
| "epoch": 2.8672, | |
| "eval_accuracy": 0.9988, | |
| "eval_f1": 0.9987998118880025, | |
| "eval_loss": 0.0046388679184019566, | |
| "eval_precision": 0.9988007388473777, | |
| "eval_recall": 0.9988, | |
| "eval_runtime": 214.8729, | |
| "eval_samples_per_second": 93.078, | |
| "eval_steps_per_second": 11.635, | |
| "step": 1792 | |
| }, | |
| { | |
| "epoch": 2.8928000000000003, | |
| "grad_norm": 0.012935050763189793, | |
| "learning_rate": 1.8133333333333335e-06, | |
| "loss": 0.0023, | |
| "step": 1808 | |
| }, | |
| { | |
| "epoch": 2.9184, | |
| "grad_norm": 1.5018342733383179, | |
| "learning_rate": 1.3866666666666666e-06, | |
| "loss": 0.0099, | |
| "step": 1824 | |
| }, | |
| { | |
| "epoch": 2.944, | |
| "grad_norm": 0.10675271600484848, | |
| "learning_rate": 9.6e-07, | |
| "loss": 0.0013, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 2.9696, | |
| "grad_norm": 0.010167909786105156, | |
| "learning_rate": 5.333333333333333e-07, | |
| "loss": 0.003, | |
| "step": 1856 | |
| }, | |
| { | |
| "epoch": 2.9952, | |
| "grad_norm": 0.014523031190037727, | |
| "learning_rate": 1.0666666666666668e-07, | |
| "loss": 0.0058, | |
| "step": 1872 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 1875, | |
| "total_flos": 1.24283947843584e+18, | |
| "train_loss": 0.03442937224805355, | |
| "train_runtime": 6631.726, | |
| "train_samples_per_second": 36.19, | |
| "train_steps_per_second": 0.283 | |
| } | |
| ], | |
| "logging_steps": 16, | |
| "max_steps": 1875, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 128, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.24283947843584e+18, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |