| { | |
| "best_metric": 0.8459659218788147, | |
| "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-900", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 2348, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.6497292518615723, | |
| "learning_rate": 0.00019914821124361162, | |
| "loss": 1.6003, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.307145357131958, | |
| "learning_rate": 0.00019829642248722317, | |
| "loss": 1.2767, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 2.9354941844940186, | |
| "learning_rate": 0.00019744463373083478, | |
| "loss": 1.2612, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.3261815309524536, | |
| "learning_rate": 0.00019659284497444633, | |
| "loss": 1.2354, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 1.5586915016174316, | |
| "learning_rate": 0.00019574105621805794, | |
| "loss": 1.0959, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.490173578262329, | |
| "learning_rate": 0.00019488926746166952, | |
| "loss": 1.0438, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 2.0831446647644043, | |
| "learning_rate": 0.0001940374787052811, | |
| "loss": 1.0841, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 2.6207799911499023, | |
| "learning_rate": 0.00019318568994889268, | |
| "loss": 1.0983, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.7383110523223877, | |
| "learning_rate": 0.00019233390119250426, | |
| "loss": 1.1775, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 2.1954941749572754, | |
| "learning_rate": 0.00019148211243611585, | |
| "loss": 1.0616, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.5817717206132879, | |
| "eval_loss": 1.0267014503479004, | |
| "eval_runtime": 39.3874, | |
| "eval_samples_per_second": 29.806, | |
| "eval_steps_per_second": 3.732, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.597124695777893, | |
| "learning_rate": 0.00019063032367972745, | |
| "loss": 1.007, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.289490818977356, | |
| "learning_rate": 0.000189778534923339, | |
| "loss": 1.0065, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 1.7088607549667358, | |
| "learning_rate": 0.00018892674616695061, | |
| "loss": 1.0204, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 2.730241537094116, | |
| "learning_rate": 0.00018807495741056217, | |
| "loss": 0.8969, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 2.9691402912139893, | |
| "learning_rate": 0.00018722316865417378, | |
| "loss": 0.953, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 2.2519712448120117, | |
| "learning_rate": 0.00018637137989778536, | |
| "loss": 0.9269, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 1.8000602722167969, | |
| "learning_rate": 0.00018551959114139694, | |
| "loss": 1.1314, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.5348334312438965, | |
| "learning_rate": 0.00018466780238500855, | |
| "loss": 0.9615, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.599938988685608, | |
| "learning_rate": 0.0001838160136286201, | |
| "loss": 0.8033, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 1.50412917137146, | |
| "learning_rate": 0.0001829642248722317, | |
| "loss": 0.9594, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.6073253833049403, | |
| "eval_loss": 0.9467767477035522, | |
| "eval_runtime": 38.8829, | |
| "eval_samples_per_second": 30.193, | |
| "eval_steps_per_second": 3.781, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 2.1896722316741943, | |
| "learning_rate": 0.0001821124361158433, | |
| "loss": 0.9217, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.9687891006469727, | |
| "learning_rate": 0.00018126064735945487, | |
| "loss": 1.0296, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 1.9628914594650269, | |
| "learning_rate": 0.00018040885860306645, | |
| "loss": 0.8122, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 2.598545789718628, | |
| "learning_rate": 0.00017955706984667803, | |
| "loss": 0.8393, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 2.2483532428741455, | |
| "learning_rate": 0.0001787052810902896, | |
| "loss": 0.9047, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 2.1274337768554688, | |
| "learning_rate": 0.0001778534923339012, | |
| "loss": 0.91, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 2.436018466949463, | |
| "learning_rate": 0.00017700170357751277, | |
| "loss": 1.0615, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.069586992263794, | |
| "learning_rate": 0.00017614991482112438, | |
| "loss": 1.0799, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.7266385555267334, | |
| "learning_rate": 0.00017529812606473594, | |
| "loss": 0.9465, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 2.0491390228271484, | |
| "learning_rate": 0.00017444633730834754, | |
| "loss": 1.1785, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_accuracy": 0.5868824531516184, | |
| "eval_loss": 0.997596025466919, | |
| "eval_runtime": 39.3421, | |
| "eval_samples_per_second": 29.841, | |
| "eval_steps_per_second": 3.736, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 1.4697805643081665, | |
| "learning_rate": 0.00017359454855195912, | |
| "loss": 1.094, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 2.369339942932129, | |
| "learning_rate": 0.0001727427597955707, | |
| "loss": 0.9398, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 2.325148344039917, | |
| "learning_rate": 0.00017189097103918229, | |
| "loss": 0.9718, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.9404678344726562, | |
| "learning_rate": 0.00017103918228279387, | |
| "loss": 0.9091, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 2.4493370056152344, | |
| "learning_rate": 0.00017018739352640547, | |
| "loss": 0.9295, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.6286579370498657, | |
| "learning_rate": 0.00016933560477001706, | |
| "loss": 1.1049, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 3.559056043624878, | |
| "learning_rate": 0.00016848381601362864, | |
| "loss": 0.9566, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.4250924587249756, | |
| "learning_rate": 0.00016763202725724022, | |
| "loss": 0.7772, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.5668089389801025, | |
| "learning_rate": 0.0001667802385008518, | |
| "loss": 0.8869, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 2.725231885910034, | |
| "learning_rate": 0.00016592844974446338, | |
| "loss": 0.865, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_accuracy": 0.6388415672913118, | |
| "eval_loss": 0.9287859201431274, | |
| "eval_runtime": 38.5489, | |
| "eval_samples_per_second": 30.455, | |
| "eval_steps_per_second": 3.813, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 2.6907713413238525, | |
| "learning_rate": 0.00016507666098807496, | |
| "loss": 0.899, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.402860164642334, | |
| "learning_rate": 0.00016422487223168654, | |
| "loss": 0.9506, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 2.749433994293213, | |
| "learning_rate": 0.00016337308347529815, | |
| "loss": 0.8529, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.92979097366333, | |
| "learning_rate": 0.0001625212947189097, | |
| "loss": 0.8695, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 2.793747901916504, | |
| "learning_rate": 0.0001616695059625213, | |
| "loss": 0.8614, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 2.483780860900879, | |
| "learning_rate": 0.0001608177172061329, | |
| "loss": 0.9176, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.7278929948806763, | |
| "learning_rate": 0.00015996592844974447, | |
| "loss": 0.9656, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 2.649017810821533, | |
| "learning_rate": 0.00015911413969335605, | |
| "loss": 0.8653, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 1.8457053899765015, | |
| "learning_rate": 0.00015826235093696763, | |
| "loss": 0.7707, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 2.824699640274048, | |
| "learning_rate": 0.00015741056218057921, | |
| "loss": 0.8494, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.651618398637138, | |
| "eval_loss": 0.8572959303855896, | |
| "eval_runtime": 38.0883, | |
| "eval_samples_per_second": 30.823, | |
| "eval_steps_per_second": 3.859, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 1.9104124307632446, | |
| "learning_rate": 0.0001565587734241908, | |
| "loss": 0.8113, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 2.2717394828796387, | |
| "learning_rate": 0.0001557069846678024, | |
| "loss": 0.9194, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.891735553741455, | |
| "learning_rate": 0.00015485519591141398, | |
| "loss": 0.9337, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 4.8229146003723145, | |
| "learning_rate": 0.00015400340715502557, | |
| "loss": 0.9033, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 2.656970977783203, | |
| "learning_rate": 0.00015315161839863715, | |
| "loss": 0.8985, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 2.2908411026000977, | |
| "learning_rate": 0.00015229982964224873, | |
| "loss": 0.8708, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 2.141950845718384, | |
| "learning_rate": 0.0001514480408858603, | |
| "loss": 0.9298, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 2.5572831630706787, | |
| "learning_rate": 0.00015059625212947192, | |
| "loss": 0.9101, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 2.2453222274780273, | |
| "learning_rate": 0.00014982964224872234, | |
| "loss": 0.8034, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 2.2874865531921387, | |
| "learning_rate": 0.00014897785349233392, | |
| "loss": 0.8151, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.6396933560477002, | |
| "eval_loss": 0.87294602394104, | |
| "eval_runtime": 38.7251, | |
| "eval_samples_per_second": 30.316, | |
| "eval_steps_per_second": 3.796, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 2.950303554534912, | |
| "learning_rate": 0.0001481260647359455, | |
| "loss": 0.7484, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 1.9773017168045044, | |
| "learning_rate": 0.00014727427597955708, | |
| "loss": 0.6572, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 2.9777700901031494, | |
| "learning_rate": 0.00014642248722316866, | |
| "loss": 0.6927, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 3.323662519454956, | |
| "learning_rate": 0.00014557069846678024, | |
| "loss": 0.5812, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 1.9647018909454346, | |
| "learning_rate": 0.00014471890971039185, | |
| "loss": 0.6166, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 3.215794563293457, | |
| "learning_rate": 0.0001438671209540034, | |
| "loss": 0.602, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 2.8758130073547363, | |
| "learning_rate": 0.000143015332197615, | |
| "loss": 0.5224, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 2.142829179763794, | |
| "learning_rate": 0.00014216354344122656, | |
| "loss": 0.5663, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 6.860159397125244, | |
| "learning_rate": 0.00014131175468483817, | |
| "loss": 0.6479, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 3.3176701068878174, | |
| "learning_rate": 0.00014045996592844975, | |
| "loss": 0.5787, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_accuracy": 0.6448040885860307, | |
| "eval_loss": 0.9067147970199585, | |
| "eval_runtime": 38.2427, | |
| "eval_samples_per_second": 30.699, | |
| "eval_steps_per_second": 3.844, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 2.322371482849121, | |
| "learning_rate": 0.00013960817717206133, | |
| "loss": 0.6849, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 1.875775933265686, | |
| "learning_rate": 0.00013875638841567291, | |
| "loss": 0.6399, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 2.0012145042419434, | |
| "learning_rate": 0.0001379045996592845, | |
| "loss": 0.725, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 2.5320353507995605, | |
| "learning_rate": 0.00013705281090289608, | |
| "loss": 0.5306, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.29856538772583, | |
| "learning_rate": 0.00013620102214650768, | |
| "loss": 0.5731, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 1.8604925870895386, | |
| "learning_rate": 0.00013534923339011926, | |
| "loss": 0.6806, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 2.6868739128112793, | |
| "learning_rate": 0.00013449744463373084, | |
| "loss": 0.5944, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 3.3680803775787354, | |
| "learning_rate": 0.00013364565587734243, | |
| "loss": 0.6412, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 2.798149824142456, | |
| "learning_rate": 0.000132793867120954, | |
| "loss": 0.5235, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 2.4862072467803955, | |
| "learning_rate": 0.00013194207836456561, | |
| "loss": 0.7768, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_accuracy": 0.6533219761499148, | |
| "eval_loss": 0.8995758295059204, | |
| "eval_runtime": 38.4107, | |
| "eval_samples_per_second": 30.564, | |
| "eval_steps_per_second": 3.827, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 3.291276216506958, | |
| "learning_rate": 0.00013109028960817717, | |
| "loss": 0.669, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.814397096633911, | |
| "learning_rate": 0.00013023850085178878, | |
| "loss": 0.5539, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 2.5982093811035156, | |
| "learning_rate": 0.00012938671209540033, | |
| "loss": 0.6565, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 3.1191565990448, | |
| "learning_rate": 0.00012853492333901194, | |
| "loss": 0.533, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 5.229197025299072, | |
| "learning_rate": 0.00012768313458262352, | |
| "loss": 0.6123, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 2.259110689163208, | |
| "learning_rate": 0.0001268313458262351, | |
| "loss": 0.5183, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 3.099496364593506, | |
| "learning_rate": 0.00012597955706984668, | |
| "loss": 0.6911, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.9909987449645996, | |
| "learning_rate": 0.00012512776831345826, | |
| "loss": 0.6671, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 3.1856462955474854, | |
| "learning_rate": 0.00012427597955706984, | |
| "loss": 0.6652, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 3.9080755710601807, | |
| "learning_rate": 0.00012342419080068145, | |
| "loss": 0.6098, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.6695059625212947, | |
| "eval_loss": 0.8459659218788147, | |
| "eval_runtime": 37.8733, | |
| "eval_samples_per_second": 30.998, | |
| "eval_steps_per_second": 3.881, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 1.7587580680847168, | |
| "learning_rate": 0.000122572402044293, | |
| "loss": 0.7362, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 2.7327494621276855, | |
| "learning_rate": 0.00012172061328790461, | |
| "loss": 0.5863, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 4.113401889801025, | |
| "learning_rate": 0.0001208688245315162, | |
| "loss": 0.8205, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 4.598094940185547, | |
| "learning_rate": 0.00012001703577512777, | |
| "loss": 0.7198, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 2.8792037963867188, | |
| "learning_rate": 0.00011916524701873937, | |
| "loss": 0.6532, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 2.949414014816284, | |
| "learning_rate": 0.00011831345826235094, | |
| "loss": 0.6783, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 2.300352096557617, | |
| "learning_rate": 0.00011746166950596253, | |
| "loss": 0.69, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 2.5100274085998535, | |
| "learning_rate": 0.00011660988074957411, | |
| "loss": 0.7028, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 2.372359275817871, | |
| "learning_rate": 0.0001157580919931857, | |
| "loss": 0.5673, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 4.268792152404785, | |
| "learning_rate": 0.00011490630323679727, | |
| "loss": 0.6251, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_accuracy": 0.6703577512776832, | |
| "eval_loss": 0.8609783053398132, | |
| "eval_runtime": 37.811, | |
| "eval_samples_per_second": 31.049, | |
| "eval_steps_per_second": 3.888, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 3.081153154373169, | |
| "learning_rate": 0.00011405451448040887, | |
| "loss": 0.7021, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 2.9631364345550537, | |
| "learning_rate": 0.0001132879045996593, | |
| "loss": 0.5469, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 3.2896649837493896, | |
| "learning_rate": 0.00011243611584327087, | |
| "loss": 0.5593, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 3.8375134468078613, | |
| "learning_rate": 0.00011158432708688246, | |
| "loss": 0.5499, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 1.5597748756408691, | |
| "learning_rate": 0.00011073253833049404, | |
| "loss": 0.5529, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 4.54299783706665, | |
| "learning_rate": 0.00010988074957410564, | |
| "loss": 0.6211, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 3.2734501361846924, | |
| "learning_rate": 0.0001090289608177172, | |
| "loss": 0.7002, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 3.7582859992980957, | |
| "learning_rate": 0.0001081771720613288, | |
| "loss": 0.7465, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 2.190544605255127, | |
| "learning_rate": 0.00010732538330494038, | |
| "loss": 0.6662, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 1.7477951049804688, | |
| "learning_rate": 0.00010647359454855197, | |
| "loss": 0.7863, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.6431005110732538, | |
| "eval_loss": 0.8668282628059387, | |
| "eval_runtime": 37.5178, | |
| "eval_samples_per_second": 31.292, | |
| "eval_steps_per_second": 3.918, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 1.9970145225524902, | |
| "learning_rate": 0.00010562180579216354, | |
| "loss": 0.5988, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 3.718055248260498, | |
| "learning_rate": 0.00010477001703577514, | |
| "loss": 0.5973, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 1.6347967386245728, | |
| "learning_rate": 0.0001039182282793867, | |
| "loss": 0.5818, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 2.3118577003479004, | |
| "learning_rate": 0.0001030664395229983, | |
| "loss": 0.5136, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 2.806833267211914, | |
| "learning_rate": 0.00010221465076660988, | |
| "loss": 0.5353, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 2.699890375137329, | |
| "learning_rate": 0.00010136286201022147, | |
| "loss": 0.5498, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 2.3461856842041016, | |
| "learning_rate": 0.00010051107325383304, | |
| "loss": 0.7181, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 2.483959436416626, | |
| "learning_rate": 9.965928449744463e-05, | |
| "loss": 0.3872, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.1393377780914307, | |
| "learning_rate": 9.880749574105622e-05, | |
| "loss": 0.292, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 1.6828927993774414, | |
| "learning_rate": 9.795570698466781e-05, | |
| "loss": 0.2595, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_accuracy": 0.6839863713798978, | |
| "eval_loss": 0.8725138902664185, | |
| "eval_runtime": 37.2408, | |
| "eval_samples_per_second": 31.525, | |
| "eval_steps_per_second": 3.947, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 1.973240852355957, | |
| "learning_rate": 9.710391822827939e-05, | |
| "loss": 0.2644, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 2.932751417160034, | |
| "learning_rate": 9.625212947189097e-05, | |
| "loss": 0.2925, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 3.356760025024414, | |
| "learning_rate": 9.540034071550255e-05, | |
| "loss": 0.312, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 1.7125446796417236, | |
| "learning_rate": 9.454855195911415e-05, | |
| "loss": 0.2277, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 1.714805006980896, | |
| "learning_rate": 9.369676320272573e-05, | |
| "loss": 0.3301, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 2.301734685897827, | |
| "learning_rate": 9.284497444633732e-05, | |
| "loss": 0.2668, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 2.4843878746032715, | |
| "learning_rate": 9.19931856899489e-05, | |
| "loss": 0.2333, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 2.9054977893829346, | |
| "learning_rate": 9.114139693356048e-05, | |
| "loss": 0.3492, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 4.664933681488037, | |
| "learning_rate": 9.028960817717206e-05, | |
| "loss": 0.3754, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 2.1164679527282715, | |
| "learning_rate": 8.943781942078366e-05, | |
| "loss": 0.2735, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.6746166950596252, | |
| "eval_loss": 0.9306557178497314, | |
| "eval_runtime": 37.0944, | |
| "eval_samples_per_second": 31.649, | |
| "eval_steps_per_second": 3.963, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 4.541740894317627, | |
| "learning_rate": 8.858603066439524e-05, | |
| "loss": 0.3835, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 3.0828359127044678, | |
| "learning_rate": 8.773424190800682e-05, | |
| "loss": 0.3189, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 2.398512363433838, | |
| "learning_rate": 8.68824531516184e-05, | |
| "loss": 0.29, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 3.069840908050537, | |
| "learning_rate": 8.603066439522998e-05, | |
| "loss": 0.288, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 5.078506946563721, | |
| "learning_rate": 8.517887563884158e-05, | |
| "loss": 0.2772, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 2.812199354171753, | |
| "learning_rate": 8.432708688245316e-05, | |
| "loss": 0.2951, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 4.542017936706543, | |
| "learning_rate": 8.347529812606474e-05, | |
| "loss": 0.2142, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 3.7486696243286133, | |
| "learning_rate": 8.262350936967632e-05, | |
| "loss": 0.257, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 3.5566983222961426, | |
| "learning_rate": 8.17717206132879e-05, | |
| "loss": 0.2816, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 1.3465384244918823, | |
| "learning_rate": 8.09199318568995e-05, | |
| "loss": 0.2429, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "eval_accuracy": 0.6354344122657581, | |
| "eval_loss": 1.0957823991775513, | |
| "eval_runtime": 37.2033, | |
| "eval_samples_per_second": 31.556, | |
| "eval_steps_per_second": 3.951, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.310131788253784, | |
| "learning_rate": 8.006814310051108e-05, | |
| "loss": 0.306, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 3.1297261714935303, | |
| "learning_rate": 7.921635434412266e-05, | |
| "loss": 0.3257, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 1.8082480430603027, | |
| "learning_rate": 7.836456558773425e-05, | |
| "loss": 0.2001, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 1.7700148820877075, | |
| "learning_rate": 7.751277683134583e-05, | |
| "loss": 0.3476, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 4.247625350952148, | |
| "learning_rate": 7.666098807495741e-05, | |
| "loss": 0.2323, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 4.059571743011475, | |
| "learning_rate": 7.5809199318569e-05, | |
| "loss": 0.3089, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.2417612075805664, | |
| "learning_rate": 7.495741056218059e-05, | |
| "loss": 0.1964, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 5.7817463874816895, | |
| "learning_rate": 7.410562180579217e-05, | |
| "loss": 0.3549, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 5.440825939178467, | |
| "learning_rate": 7.325383304940375e-05, | |
| "loss": 0.3085, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 4.482067108154297, | |
| "learning_rate": 7.240204429301533e-05, | |
| "loss": 0.3224, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_accuracy": 0.6686541737649063, | |
| "eval_loss": 1.0305246114730835, | |
| "eval_runtime": 37.1181, | |
| "eval_samples_per_second": 31.629, | |
| "eval_steps_per_second": 3.96, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 2.1568057537078857, | |
| "learning_rate": 7.155025553662692e-05, | |
| "loss": 0.1612, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 1.293427586555481, | |
| "learning_rate": 7.06984667802385e-05, | |
| "loss": 0.3217, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 4.301244258880615, | |
| "learning_rate": 6.984667802385009e-05, | |
| "loss": 0.2378, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 1.6040468215942383, | |
| "learning_rate": 6.899488926746167e-05, | |
| "loss": 0.2801, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.7993047833442688, | |
| "learning_rate": 6.814310051107326e-05, | |
| "loss": 0.2637, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 4.865533828735352, | |
| "learning_rate": 6.729131175468484e-05, | |
| "loss": 0.3441, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 1.7501546144485474, | |
| "learning_rate": 6.643952299829642e-05, | |
| "loss": 0.2523, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 1.331475019454956, | |
| "learning_rate": 6.5587734241908e-05, | |
| "loss": 0.2127, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "grad_norm": 3.352147102355957, | |
| "learning_rate": 6.473594548551958e-05, | |
| "loss": 0.3432, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 0.3470512330532074, | |
| "learning_rate": 6.388415672913118e-05, | |
| "loss": 0.1602, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_accuracy": 0.6746166950596252, | |
| "eval_loss": 1.0072139501571655, | |
| "eval_runtime": 37.0019, | |
| "eval_samples_per_second": 31.728, | |
| "eval_steps_per_second": 3.973, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 3.1594250202178955, | |
| "learning_rate": 6.303236797274277e-05, | |
| "loss": 0.1929, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 4.477923393249512, | |
| "learning_rate": 6.218057921635435e-05, | |
| "loss": 0.2696, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 3.042938232421875, | |
| "learning_rate": 6.132879045996594e-05, | |
| "loss": 0.2527, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 0.8534514904022217, | |
| "learning_rate": 6.0477001703577516e-05, | |
| "loss": 0.1727, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 2.2307116985321045, | |
| "learning_rate": 5.9625212947189104e-05, | |
| "loss": 0.3178, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 3.302003860473633, | |
| "learning_rate": 5.8773424190800684e-05, | |
| "loss": 0.2973, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 5.320656776428223, | |
| "learning_rate": 5.792163543441227e-05, | |
| "loss": 0.2715, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "grad_norm": 3.923163414001465, | |
| "learning_rate": 5.706984667802385e-05, | |
| "loss": 0.1991, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 7.479254245758057, | |
| "learning_rate": 5.6218057921635434e-05, | |
| "loss": 0.321, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 2.2710225582122803, | |
| "learning_rate": 5.536626916524702e-05, | |
| "loss": 0.2042, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.6788756388415673, | |
| "eval_loss": 1.0971218347549438, | |
| "eval_runtime": 36.9173, | |
| "eval_samples_per_second": 31.801, | |
| "eval_steps_per_second": 3.982, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 2.7610058784484863, | |
| "learning_rate": 5.45144804088586e-05, | |
| "loss": 0.3396, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "grad_norm": 2.2475104331970215, | |
| "learning_rate": 5.366269165247019e-05, | |
| "loss": 0.266, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 4.55673885345459, | |
| "learning_rate": 5.281090289608177e-05, | |
| "loss": 0.341, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 4.0248260498046875, | |
| "learning_rate": 5.195911413969335e-05, | |
| "loss": 0.2005, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 4.798257827758789, | |
| "learning_rate": 5.110732538330494e-05, | |
| "loss": 0.2615, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.2967402935028076, | |
| "learning_rate": 5.025553662691652e-05, | |
| "loss": 0.1966, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 5.774517059326172, | |
| "learning_rate": 4.940374787052811e-05, | |
| "loss": 0.1141, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 1.7739803791046143, | |
| "learning_rate": 4.8551959114139695e-05, | |
| "loss": 0.0671, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 0.8837150931358337, | |
| "learning_rate": 4.7700170357751276e-05, | |
| "loss": 0.0835, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 1.7833037376403809, | |
| "learning_rate": 4.6848381601362864e-05, | |
| "loss": 0.0604, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "eval_accuracy": 0.6916524701873935, | |
| "eval_loss": 1.0816737413406372, | |
| "eval_runtime": 36.8222, | |
| "eval_samples_per_second": 31.883, | |
| "eval_steps_per_second": 3.992, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 0.34585830569267273, | |
| "learning_rate": 4.599659284497445e-05, | |
| "loss": 0.092, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 0.7962571382522583, | |
| "learning_rate": 4.514480408858603e-05, | |
| "loss": 0.0587, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.16402888298034668, | |
| "learning_rate": 4.429301533219762e-05, | |
| "loss": 0.0547, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 0.624047040939331, | |
| "learning_rate": 4.34412265758092e-05, | |
| "loss": 0.0954, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 0.4253842532634735, | |
| "learning_rate": 4.258943781942079e-05, | |
| "loss": 0.0567, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 0.1523701399564743, | |
| "learning_rate": 4.173764906303237e-05, | |
| "loss": 0.0413, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 4.592818260192871, | |
| "learning_rate": 4.088586030664395e-05, | |
| "loss": 0.0968, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.4066344499588013, | |
| "learning_rate": 4.003407155025554e-05, | |
| "loss": 0.1454, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 2.1996095180511475, | |
| "learning_rate": 3.9182282793867125e-05, | |
| "loss": 0.1128, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "grad_norm": 0.102027028799057, | |
| "learning_rate": 3.8330494037478706e-05, | |
| "loss": 0.0716, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_accuracy": 0.692504258943782, | |
| "eval_loss": 1.1307132244110107, | |
| "eval_runtime": 37.0403, | |
| "eval_samples_per_second": 31.695, | |
| "eval_steps_per_second": 3.969, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 1.6857343912124634, | |
| "learning_rate": 3.7478705281090294e-05, | |
| "loss": 0.04, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "grad_norm": 1.2973403930664062, | |
| "learning_rate": 3.6626916524701875e-05, | |
| "loss": 0.0403, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "grad_norm": 0.41860514879226685, | |
| "learning_rate": 3.577512776831346e-05, | |
| "loss": 0.0642, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 0.5436795353889465, | |
| "learning_rate": 3.492333901192504e-05, | |
| "loss": 0.0836, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 0.21996204555034637, | |
| "learning_rate": 3.407155025553663e-05, | |
| "loss": 0.0406, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "grad_norm": 0.14845231175422668, | |
| "learning_rate": 3.321976149914821e-05, | |
| "loss": 0.0385, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 3.531405448913574, | |
| "learning_rate": 3.236797274275979e-05, | |
| "loss": 0.0824, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "grad_norm": 0.07682117819786072, | |
| "learning_rate": 3.151618398637139e-05, | |
| "loss": 0.0717, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 0.07611515372991562, | |
| "learning_rate": 3.066439522998297e-05, | |
| "loss": 0.0572, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "grad_norm": 0.6266534328460693, | |
| "learning_rate": 2.9812606473594552e-05, | |
| "loss": 0.0822, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "eval_accuracy": 0.692504258943782, | |
| "eval_loss": 1.1826940774917603, | |
| "eval_runtime": 37.1369, | |
| "eval_samples_per_second": 31.613, | |
| "eval_steps_per_second": 3.958, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "grad_norm": 0.1280030608177185, | |
| "learning_rate": 2.8960817717206136e-05, | |
| "loss": 0.0244, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 0.07406999170780182, | |
| "learning_rate": 2.8109028960817717e-05, | |
| "loss": 0.0574, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "grad_norm": 5.587332248687744, | |
| "learning_rate": 2.72572402044293e-05, | |
| "loss": 0.0352, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 2.2010979652404785, | |
| "learning_rate": 2.6405451448040885e-05, | |
| "loss": 0.0789, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "grad_norm": 2.9271368980407715, | |
| "learning_rate": 2.555366269165247e-05, | |
| "loss": 0.082, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "grad_norm": 0.05890679359436035, | |
| "learning_rate": 2.4701873935264054e-05, | |
| "loss": 0.0769, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "grad_norm": 0.7043523192405701, | |
| "learning_rate": 2.3850085178875638e-05, | |
| "loss": 0.0819, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 0.12047506123781204, | |
| "learning_rate": 2.2998296422487226e-05, | |
| "loss": 0.0195, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "grad_norm": 0.1116802990436554, | |
| "learning_rate": 2.214650766609881e-05, | |
| "loss": 0.0159, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "grad_norm": 0.09187493473291397, | |
| "learning_rate": 2.1294718909710394e-05, | |
| "loss": 0.0889, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "eval_accuracy": 0.6933560477001703, | |
| "eval_loss": 1.2423571348190308, | |
| "eval_runtime": 37.3059, | |
| "eval_samples_per_second": 31.47, | |
| "eval_steps_per_second": 3.94, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "grad_norm": 4.332376956939697, | |
| "learning_rate": 2.0442930153321975e-05, | |
| "loss": 0.0939, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 0.13916102051734924, | |
| "learning_rate": 1.9591141396933563e-05, | |
| "loss": 0.0933, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "grad_norm": 7.690703392028809, | |
| "learning_rate": 1.8739352640545147e-05, | |
| "loss": 0.0496, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 2.5700595378875732, | |
| "learning_rate": 1.788756388415673e-05, | |
| "loss": 0.0782, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "grad_norm": 0.20934216678142548, | |
| "learning_rate": 1.7035775127768315e-05, | |
| "loss": 0.0606, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 1.2959486246109009, | |
| "learning_rate": 1.6183986371379896e-05, | |
| "loss": 0.0601, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 0.2652721405029297, | |
| "learning_rate": 1.5332197614991484e-05, | |
| "loss": 0.062, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "grad_norm": 0.48360127210617065, | |
| "learning_rate": 1.4480408858603068e-05, | |
| "loss": 0.054, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "grad_norm": 3.1118693351745605, | |
| "learning_rate": 1.362862010221465e-05, | |
| "loss": 0.0989, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.9077383279800415, | |
| "learning_rate": 1.2776831345826235e-05, | |
| "loss": 0.0855, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "eval_accuracy": 0.6899488926746167, | |
| "eval_loss": 1.2667156457901, | |
| "eval_runtime": 36.8511, | |
| "eval_samples_per_second": 31.858, | |
| "eval_steps_per_second": 3.989, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.13304296135902405, | |
| "learning_rate": 1.1925042589437819e-05, | |
| "loss": 0.0675, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "grad_norm": 1.3241567611694336, | |
| "learning_rate": 1.1073253833049405e-05, | |
| "loss": 0.0753, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.2818525731563568, | |
| "learning_rate": 1.0221465076660987e-05, | |
| "loss": 0.0998, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "grad_norm": 7.136697292327881, | |
| "learning_rate": 9.369676320272573e-06, | |
| "loss": 0.0314, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 1.372044324874878, | |
| "learning_rate": 8.517887563884158e-06, | |
| "loss": 0.0768, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "grad_norm": 6.264348983764648, | |
| "learning_rate": 7.666098807495742e-06, | |
| "loss": 0.1516, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 0.1342085599899292, | |
| "learning_rate": 6.814310051107325e-06, | |
| "loss": 0.0812, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "grad_norm": 0.7664629220962524, | |
| "learning_rate": 5.9625212947189095e-06, | |
| "loss": 0.0474, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 4.264090538024902, | |
| "learning_rate": 5.110732538330494e-06, | |
| "loss": 0.0903, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.07316776365041733, | |
| "learning_rate": 4.258943781942079e-06, | |
| "loss": 0.0682, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "eval_accuracy": 0.6950596252129472, | |
| "eval_loss": 1.2470241785049438, | |
| "eval_runtime": 37.0027, | |
| "eval_samples_per_second": 31.727, | |
| "eval_steps_per_second": 3.973, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "grad_norm": 1.477973222732544, | |
| "learning_rate": 3.4071550255536626e-06, | |
| "loss": 0.0587, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 1.249779224395752, | |
| "learning_rate": 2.555366269165247e-06, | |
| "loss": 0.0546, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "grad_norm": 1.9763495922088623, | |
| "learning_rate": 1.7035775127768313e-06, | |
| "loss": 0.0539, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "grad_norm": 0.11824575811624527, | |
| "learning_rate": 8.517887563884157e-07, | |
| "loss": 0.0322, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 2348, | |
| "total_flos": 2.910419581971751e+18, | |
| "train_loss": 0.4888155373286145, | |
| "train_runtime": 2894.9609, | |
| "train_samples_per_second": 12.973, | |
| "train_steps_per_second": 0.811 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2348, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "total_flos": 2.910419581971751e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |