| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.24064171122994651, | |
| "eval_steps": 10, | |
| "global_step": 540, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004456327985739751, | |
| "grad_norm": 1.5512940883636475, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.4724, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004456327985739751, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.34007999300956726, | |
| "eval_runtime": 549.9948, | |
| "eval_samples_per_second": 8.16, | |
| "eval_steps_per_second": 2.04, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "grad_norm": 1.4623041152954102, | |
| "learning_rate": 9.636363636363637e-05, | |
| "loss": 0.2715, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.39523470401763916, | |
| "eval_runtime": 543.4137, | |
| "eval_samples_per_second": 8.259, | |
| "eval_steps_per_second": 2.065, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "grad_norm": 0.6407843828201294, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.3566, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.38253363966941833, | |
| "eval_runtime": 536.8913, | |
| "eval_samples_per_second": 8.359, | |
| "eval_steps_per_second": 2.09, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "grad_norm": 1.9206137657165527, | |
| "learning_rate": 9.272727272727273e-05, | |
| "loss": 0.2892, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.36841902136802673, | |
| "eval_runtime": 539.4459, | |
| "eval_samples_per_second": 8.32, | |
| "eval_steps_per_second": 2.08, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "grad_norm": 1.0926023721694946, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.3157, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3572486639022827, | |
| "eval_runtime": 542.4121, | |
| "eval_samples_per_second": 8.274, | |
| "eval_steps_per_second": 2.069, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "grad_norm": 3.161236524581909, | |
| "learning_rate": 8.90909090909091e-05, | |
| "loss": 0.3792, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3476051688194275, | |
| "eval_runtime": 538.7398, | |
| "eval_samples_per_second": 8.331, | |
| "eval_steps_per_second": 2.083, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "grad_norm": 1.0513850450515747, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.3938, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.317058265209198, | |
| "eval_runtime": 540.0617, | |
| "eval_samples_per_second": 8.31, | |
| "eval_steps_per_second": 2.078, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "grad_norm": 5.6621479988098145, | |
| "learning_rate": 8.545454545454545e-05, | |
| "loss": 0.3962, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2908115088939667, | |
| "eval_runtime": 541.5124, | |
| "eval_samples_per_second": 8.288, | |
| "eval_steps_per_second": 2.072, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "grad_norm": 11.542706489562988, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.3536, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2727506160736084, | |
| "eval_runtime": 542.0266, | |
| "eval_samples_per_second": 8.28, | |
| "eval_steps_per_second": 2.07, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "grad_norm": 0.7968679070472717, | |
| "learning_rate": 8.181818181818183e-05, | |
| "loss": 0.2338, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "eval_accuracy": 0.898172914981842, | |
| "eval_loss": 0.21359723806381226, | |
| "eval_runtime": 550.439, | |
| "eval_samples_per_second": 8.153, | |
| "eval_steps_per_second": 2.038, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "grad_norm": 13.235169410705566, | |
| "learning_rate": 8e-05, | |
| "loss": 0.2591, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "eval_accuracy": 0.9035205245018005, | |
| "eval_loss": 0.17823095619678497, | |
| "eval_runtime": 543.3444, | |
| "eval_samples_per_second": 8.26, | |
| "eval_steps_per_second": 2.065, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "grad_norm": 3.1415486335754395, | |
| "learning_rate": 7.818181818181818e-05, | |
| "loss": 0.261, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "eval_accuracy": 0.8725489974021912, | |
| "eval_loss": 0.23705999553203583, | |
| "eval_runtime": 550.4034, | |
| "eval_samples_per_second": 8.154, | |
| "eval_steps_per_second": 2.039, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "grad_norm": 0.12994514405727386, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.2626, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.4889169931411743, | |
| "eval_runtime": 545.9677, | |
| "eval_samples_per_second": 8.22, | |
| "eval_steps_per_second": 2.055, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "grad_norm": 0.20874539017677307, | |
| "learning_rate": 7.454545454545455e-05, | |
| "loss": 0.3156, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "eval_accuracy": 0.9021835923194885, | |
| "eval_loss": 0.21060487627983093, | |
| "eval_runtime": 543.8063, | |
| "eval_samples_per_second": 8.253, | |
| "eval_steps_per_second": 2.063, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "grad_norm": 1.1314120292663574, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.3342, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "eval_accuracy": 0.9217914342880249, | |
| "eval_loss": 0.19053570926189423, | |
| "eval_runtime": 543.9184, | |
| "eval_samples_per_second": 8.251, | |
| "eval_steps_per_second": 2.063, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "grad_norm": 5.4050092697143555, | |
| "learning_rate": 7.090909090909092e-05, | |
| "loss": 0.2658, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "eval_accuracy": 0.9498662948608398, | |
| "eval_loss": 0.12423694878816605, | |
| "eval_runtime": 542.375, | |
| "eval_samples_per_second": 8.275, | |
| "eval_steps_per_second": 2.069, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "grad_norm": 6.958705425262451, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.2162, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "eval_accuracy": 0.9211229681968689, | |
| "eval_loss": 0.1585531085729599, | |
| "eval_runtime": 550.3192, | |
| "eval_samples_per_second": 8.155, | |
| "eval_steps_per_second": 2.039, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "grad_norm": 6.347085475921631, | |
| "learning_rate": 6.727272727272727e-05, | |
| "loss": 0.2457, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "eval_accuracy": 0.8422459959983826, | |
| "eval_loss": 0.31698858737945557, | |
| "eval_runtime": 550.4639, | |
| "eval_samples_per_second": 8.153, | |
| "eval_steps_per_second": 2.038, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "grad_norm": 0.05648183450102806, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.178, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "eval_accuracy": 0.9663547277450562, | |
| "eval_loss": 0.08927226811647415, | |
| "eval_runtime": 548.6704, | |
| "eval_samples_per_second": 8.18, | |
| "eval_steps_per_second": 2.045, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "grad_norm": 0.10795829445123672, | |
| "learning_rate": 6.363636363636364e-05, | |
| "loss": 0.2864, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "eval_accuracy": 0.9079768061637878, | |
| "eval_loss": 0.1990886628627777, | |
| "eval_runtime": 547.327, | |
| "eval_samples_per_second": 8.2, | |
| "eval_steps_per_second": 2.05, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "grad_norm": 0.10935252159833908, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.0852, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "eval_accuracy": 0.9777183532714844, | |
| "eval_loss": 0.06789236515760422, | |
| "eval_runtime": 546.4384, | |
| "eval_samples_per_second": 8.213, | |
| "eval_steps_per_second": 2.053, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "grad_norm": 0.724420428276062, | |
| "learning_rate": 6e-05, | |
| "loss": 0.2227, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "eval_accuracy": 0.9676916003227234, | |
| "eval_loss": 0.08267948776483536, | |
| "eval_runtime": 543.5229, | |
| "eval_samples_per_second": 8.257, | |
| "eval_steps_per_second": 2.064, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "grad_norm": 3.5009875297546387, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.0894, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "eval_accuracy": 0.9799465537071228, | |
| "eval_loss": 0.05659499019384384, | |
| "eval_runtime": 547.0317, | |
| "eval_samples_per_second": 8.204, | |
| "eval_steps_per_second": 2.051, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "grad_norm": 13.977231979370117, | |
| "learning_rate": 5.636363636363636e-05, | |
| "loss": 0.1766, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "eval_accuracy": 0.9275846481323242, | |
| "eval_loss": 0.17180828750133514, | |
| "eval_runtime": 554.5064, | |
| "eval_samples_per_second": 8.094, | |
| "eval_steps_per_second": 2.023, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "grad_norm": 1.5852386951446533, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.1133, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "eval_accuracy": 0.9625668525695801, | |
| "eval_loss": 0.09958070516586304, | |
| "eval_runtime": 546.2195, | |
| "eval_samples_per_second": 8.216, | |
| "eval_steps_per_second": 2.054, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "grad_norm": 4.984052658081055, | |
| "learning_rate": 5.272727272727272e-05, | |
| "loss": 0.1581, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "eval_accuracy": 0.9616755843162537, | |
| "eval_loss": 0.09552862495183945, | |
| "eval_runtime": 542.5097, | |
| "eval_samples_per_second": 8.273, | |
| "eval_steps_per_second": 2.068, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "grad_norm": 0.06418484449386597, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.1164, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "eval_accuracy": 0.9962121248245239, | |
| "eval_loss": 0.017139658331871033, | |
| "eval_runtime": 545.1464, | |
| "eval_samples_per_second": 8.233, | |
| "eval_steps_per_second": 2.058, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "grad_norm": 1.2460925579071045, | |
| "learning_rate": 4.909090909090909e-05, | |
| "loss": 0.0199, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "eval_accuracy": 0.9884135723114014, | |
| "eval_loss": 0.04466737061738968, | |
| "eval_runtime": 545.2406, | |
| "eval_samples_per_second": 8.231, | |
| "eval_steps_per_second": 2.058, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "grad_norm": 0.021525170654058456, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.0358, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "eval_accuracy": 0.9625668525695801, | |
| "eval_loss": 0.08889108896255493, | |
| "eval_runtime": 551.6453, | |
| "eval_samples_per_second": 8.136, | |
| "eval_steps_per_second": 2.034, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "grad_norm": 10.565823554992676, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.0134, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "eval_accuracy": 0.9607843160629272, | |
| "eval_loss": 0.11075662821531296, | |
| "eval_runtime": 546.8222, | |
| "eval_samples_per_second": 8.207, | |
| "eval_steps_per_second": 2.052, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "grad_norm": 2.769604444503784, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.0085, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "eval_accuracy": 0.9204545617103577, | |
| "eval_loss": 0.24255433678627014, | |
| "eval_runtime": 551.4065, | |
| "eval_samples_per_second": 8.139, | |
| "eval_steps_per_second": 2.035, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "grad_norm": 0.027551617473363876, | |
| "learning_rate": 4.181818181818182e-05, | |
| "loss": 0.0691, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "eval_accuracy": 0.9496434926986694, | |
| "eval_loss": 0.15619446337223053, | |
| "eval_runtime": 545.5513, | |
| "eval_samples_per_second": 8.227, | |
| "eval_steps_per_second": 2.057, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 0.010140771977603436, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2242, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "eval_accuracy": 0.9968805909156799, | |
| "eval_loss": 0.01314464956521988, | |
| "eval_runtime": 545.9537, | |
| "eval_samples_per_second": 8.22, | |
| "eval_steps_per_second": 2.055, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 0.011543634347617626, | |
| "learning_rate": 3.818181818181819e-05, | |
| "loss": 0.1593, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "eval_accuracy": 0.9890819787979126, | |
| "eval_loss": 0.03110310062766075, | |
| "eval_runtime": 547.8599, | |
| "eval_samples_per_second": 8.192, | |
| "eval_steps_per_second": 2.048, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "grad_norm": 0.06019105017185211, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0065, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "eval_accuracy": 0.9643493890762329, | |
| "eval_loss": 0.11227227747440338, | |
| "eval_runtime": 546.9694, | |
| "eval_samples_per_second": 8.205, | |
| "eval_steps_per_second": 2.051, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "grad_norm": 0.028353577479720116, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 0.0626, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "eval_accuracy": 0.9817290306091309, | |
| "eval_loss": 0.061965711414813995, | |
| "eval_runtime": 552.3776, | |
| "eval_samples_per_second": 8.125, | |
| "eval_steps_per_second": 2.031, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "grad_norm": 0.4727762043476105, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.0281, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "eval_accuracy": 0.9596702456474304, | |
| "eval_loss": 0.12898869812488556, | |
| "eval_runtime": 546.7154, | |
| "eval_samples_per_second": 8.209, | |
| "eval_steps_per_second": 2.052, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "grad_norm": 0.013356081210076809, | |
| "learning_rate": 3.090909090909091e-05, | |
| "loss": 0.0189, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "eval_accuracy": 0.991310179233551, | |
| "eval_loss": 0.02557438611984253, | |
| "eval_runtime": 546.8803, | |
| "eval_samples_per_second": 8.207, | |
| "eval_steps_per_second": 2.052, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "grad_norm": 0.056645121425390244, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.1307, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "eval_accuracy": 0.9783868193626404, | |
| "eval_loss": 0.07275046408176422, | |
| "eval_runtime": 548.2067, | |
| "eval_samples_per_second": 8.187, | |
| "eval_steps_per_second": 2.047, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "grad_norm": 0.0406530387699604, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.0061, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "eval_accuracy": 0.9471924901008606, | |
| "eval_loss": 0.17399340867996216, | |
| "eval_runtime": 542.8312, | |
| "eval_samples_per_second": 8.268, | |
| "eval_steps_per_second": 2.067, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "grad_norm": 0.07212503999471664, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.0739, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "eval_accuracy": 0.9500890970230103, | |
| "eval_loss": 0.16757053136825562, | |
| "eval_runtime": 542.9814, | |
| "eval_samples_per_second": 8.265, | |
| "eval_steps_per_second": 2.066, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "grad_norm": 0.10422785580158234, | |
| "learning_rate": 2.3636363636363637e-05, | |
| "loss": 0.0028, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "eval_accuracy": 0.9783868193626404, | |
| "eval_loss": 0.07298260927200317, | |
| "eval_runtime": 539.9794, | |
| "eval_samples_per_second": 8.311, | |
| "eval_steps_per_second": 2.078, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "grad_norm": 0.007547458633780479, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.0011, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "eval_accuracy": 0.977495551109314, | |
| "eval_loss": 0.07658497989177704, | |
| "eval_runtime": 549.3337, | |
| "eval_samples_per_second": 8.17, | |
| "eval_steps_per_second": 2.042, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "grad_norm": 0.007210019510239363, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0019, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "eval_accuracy": 0.9636809229850769, | |
| "eval_loss": 0.11826927214860916, | |
| "eval_runtime": 543.2619, | |
| "eval_samples_per_second": 8.261, | |
| "eval_steps_per_second": 2.065, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "grad_norm": 0.013209226541221142, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0388, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "eval_accuracy": 0.9576649069786072, | |
| "eval_loss": 0.14545659720897675, | |
| "eval_runtime": 544.4489, | |
| "eval_samples_per_second": 8.243, | |
| "eval_steps_per_second": 2.061, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "grad_norm": 0.007217989303171635, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 0.0041, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "eval_accuracy": 0.9523172974586487, | |
| "eval_loss": 0.16972462832927704, | |
| "eval_runtime": 545.2894, | |
| "eval_samples_per_second": 8.23, | |
| "eval_steps_per_second": 2.058, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "grad_norm": 0.8174325227737427, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.0064, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "eval_accuracy": 0.9509803652763367, | |
| "eval_loss": 0.17997007071971893, | |
| "eval_runtime": 543.7245, | |
| "eval_samples_per_second": 8.254, | |
| "eval_steps_per_second": 2.064, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "grad_norm": 0.032331835478544235, | |
| "learning_rate": 1.2727272727272727e-05, | |
| "loss": 0.0008, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "eval_accuracy": 0.9610071182250977, | |
| "eval_loss": 0.1397457718849182, | |
| "eval_runtime": 546.6811, | |
| "eval_samples_per_second": 8.21, | |
| "eval_steps_per_second": 2.052, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "grad_norm": 6.01271915435791, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.0688, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "eval_accuracy": 0.9694741368293762, | |
| "eval_loss": 0.10627125203609467, | |
| "eval_runtime": 544.0334, | |
| "eval_samples_per_second": 8.249, | |
| "eval_steps_per_second": 2.062, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "grad_norm": 0.004174210596829653, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.0008, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "eval_accuracy": 0.9844028353691101, | |
| "eval_loss": 0.05713631212711334, | |
| "eval_runtime": 551.9061, | |
| "eval_samples_per_second": 8.132, | |
| "eval_steps_per_second": 2.033, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 0.006048465613275766, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.0018, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "eval_accuracy": 0.9870766401290894, | |
| "eval_loss": 0.048317644745111465, | |
| "eval_runtime": 541.5344, | |
| "eval_samples_per_second": 8.288, | |
| "eval_steps_per_second": 2.072, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "grad_norm": 13.882183074951172, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 0.1531, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "eval_accuracy": 0.9870766401290894, | |
| "eval_loss": 0.04793470725417137, | |
| "eval_runtime": 540.1145, | |
| "eval_samples_per_second": 8.309, | |
| "eval_steps_per_second": 2.077, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "grad_norm": 0.21645870804786682, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.0027, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "eval_accuracy": 0.9848484992980957, | |
| "eval_loss": 0.055766720324754715, | |
| "eval_runtime": 545.2691, | |
| "eval_samples_per_second": 8.231, | |
| "eval_steps_per_second": 2.058, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "grad_norm": 0.27186936140060425, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 0.001, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "eval_accuracy": 0.9844028353691101, | |
| "eval_loss": 0.059176359325647354, | |
| "eval_runtime": 537.6305, | |
| "eval_samples_per_second": 8.348, | |
| "eval_steps_per_second": 2.087, | |
| "step": 540 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.207949973692707e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |