| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9998971722365039, | |
| "global_step": 4862, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0, | |
| "loss": 11.981, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 3.9970117109141705e-05, | |
| "loss": 12.0789, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 9.331893267009234e-05, | |
| "loss": 10.7133, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00011407670594843083, | |
| "loss": 8.7339, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00012721122651399258, | |
| "loss": 8.301, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00013684136855727938, | |
| "loss": 8.1964, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_accuracy": 0.10310832025117739, | |
| "eval_loss": 8.203125, | |
| "eval_runtime": 6.5764, | |
| "eval_samples_per_second": 5.322, | |
| "eval_steps_per_second": 0.304, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00014444862339428802, | |
| "loss": 8.0553, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00015073705430110066, | |
| "loss": 7.9436, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.00015609707636042195, | |
| "loss": 7.8368, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.00016076788727202945, | |
| "loss": 7.7333, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.00016490670495758757, | |
| "loss": 7.6139, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_accuracy": 0.12459794174079888, | |
| "eval_loss": 7.81640625, | |
| "eval_runtime": 6.6059, | |
| "eval_samples_per_second": 5.298, | |
| "eval_steps_per_second": 0.303, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0001686224178807056, | |
| "loss": 7.4892, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.000171993565594773, | |
| "loss": 7.3256, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00017507866443784335, | |
| "loss": 7.1827, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0001779224840062419, | |
| "loss": 6.9698, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00018056004207494319, | |
| "loss": 6.8162, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.1567556253270539, | |
| "eval_loss": 7.08203125, | |
| "eval_runtime": 6.5908, | |
| "eval_samples_per_second": 5.31, | |
| "eval_steps_per_second": 0.303, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00018301924610008189, | |
| "loss": 6.6293, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00018532269677939782, | |
| "loss": 6.4114, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00018748895370481112, | |
| "loss": 6.2911, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00018953344483335556, | |
| "loss": 6.1047, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00019146913367833817, | |
| "loss": 5.9957, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_accuracy": 0.19811616954474098, | |
| "eval_loss": 6.4296875, | |
| "eval_runtime": 6.5897, | |
| "eval_samples_per_second": 5.311, | |
| "eval_steps_per_second": 0.304, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00019330701776944063, | |
| "loss": 5.8281, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019505650713185044, | |
| "loss": 5.6927, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019672571585424665, | |
| "loss": 5.5564, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019832168964685297, | |
| "loss": 5.3813, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0001998505855457085, | |
| "loss": 5.2496, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_accuracy": 0.24055817198674342, | |
| "eval_loss": 5.8203125, | |
| "eval_runtime": 6.6013, | |
| "eval_samples_per_second": 5.302, | |
| "eval_steps_per_second": 0.303, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.0001996535296665223, | |
| "loss": 5.13, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0001992204417496752, | |
| "loss": 5.0354, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019878735383282807, | |
| "loss": 4.9021, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019835426591598097, | |
| "loss": 4.8181, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019792117799913384, | |
| "loss": 4.6993, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.2641758241758242, | |
| "eval_loss": 5.41796875, | |
| "eval_runtime": 6.6009, | |
| "eval_samples_per_second": 5.302, | |
| "eval_steps_per_second": 0.303, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0001974880900822867, | |
| "loss": 4.6761, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019705500216543958, | |
| "loss": 4.5908, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019662191424859245, | |
| "loss": 4.5301, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019618882633174535, | |
| "loss": 4.4729, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019575573841489822, | |
| "loss": 4.3928, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_accuracy": 0.27927437641723357, | |
| "eval_loss": 5.14453125, | |
| "eval_runtime": 6.6101, | |
| "eval_samples_per_second": 5.295, | |
| "eval_steps_per_second": 0.303, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019532265049805112, | |
| "loss": 4.4083, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.000194889562581204, | |
| "loss": 4.336, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0001944564746643569, | |
| "loss": 4.2714, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019402338674750976, | |
| "loss": 4.2124, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019359029883066263, | |
| "loss": 4.2395, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.2894470608756323, | |
| "eval_loss": 4.96875, | |
| "eval_runtime": 6.5945, | |
| "eval_samples_per_second": 5.307, | |
| "eval_steps_per_second": 0.303, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0001931572109138155, | |
| "loss": 4.1867, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0001927241229969684, | |
| "loss": 4.1687, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019229103508012127, | |
| "loss": 4.1027, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019185794716327414, | |
| "loss": 4.1233, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019142485924642704, | |
| "loss": 4.0781, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.29638932496075354, | |
| "eval_loss": 4.8515625, | |
| "eval_runtime": 6.5996, | |
| "eval_samples_per_second": 5.303, | |
| "eval_steps_per_second": 0.303, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0001909917713295799, | |
| "loss": 4.0855, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0001905586834127328, | |
| "loss": 4.0859, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00019012559549588568, | |
| "loss": 4.0124, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00018969250757903855, | |
| "loss": 4.0151, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00018925941966219142, | |
| "loss": 4.0409, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.30181057038199893, | |
| "eval_loss": 4.76953125, | |
| "eval_runtime": 6.6117, | |
| "eval_samples_per_second": 5.294, | |
| "eval_steps_per_second": 0.302, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00018882633174534431, | |
| "loss": 3.9912, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00018839324382849718, | |
| "loss": 3.9383, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00018796015591165008, | |
| "loss": 3.9764, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00018752706799480295, | |
| "loss": 3.9809, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00018709398007795585, | |
| "loss": 3.9178, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_accuracy": 0.30575963718820864, | |
| "eval_loss": 4.703125, | |
| "eval_runtime": 6.6265, | |
| "eval_samples_per_second": 5.282, | |
| "eval_steps_per_second": 0.302, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00018666089216110872, | |
| "loss": 3.9073, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0001862278042442616, | |
| "loss": 3.9459, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00018579471632741446, | |
| "loss": 3.9535, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00018536162841056733, | |
| "loss": 3.8982, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00018492854049372023, | |
| "loss": 3.834, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.3082574568288854, | |
| "eval_loss": 4.65625, | |
| "eval_runtime": 6.6135, | |
| "eval_samples_per_second": 5.292, | |
| "eval_steps_per_second": 0.302, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.0001844954525768731, | |
| "loss": 3.8759, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.000184062364660026, | |
| "loss": 3.8835, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00018362927674317887, | |
| "loss": 3.9003, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00018319618882633177, | |
| "loss": 3.8538, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00018276310090948464, | |
| "loss": 3.8316, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_accuracy": 0.30981336124193265, | |
| "eval_loss": 4.625, | |
| "eval_runtime": 6.6039, | |
| "eval_samples_per_second": 5.3, | |
| "eval_steps_per_second": 0.303, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00018233001299263754, | |
| "loss": 3.9085, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00018189692507579038, | |
| "loss": 3.7825, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.14, | |