| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 828, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.036231884057971016, | |
| "grad_norm": 10.94329153798689, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0543, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07246376811594203, | |
| "grad_norm": 6.69746195176836, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9358, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10869565217391304, | |
| "grad_norm": 1.4693988083101803, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8947, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14492753623188406, | |
| "grad_norm": 1.1956268436230149, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8568, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18115942028985507, | |
| "grad_norm": 1.161881545591458, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8308, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 1.0465718565340598, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8168, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2536231884057971, | |
| "grad_norm": 0.8500018086045097, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8027, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2898550724637681, | |
| "grad_norm": 0.7729963066093143, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7872, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.32608695652173914, | |
| "grad_norm": 0.9101222071301408, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7819, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.36231884057971014, | |
| "grad_norm": 1.0288861721148082, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7754, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.39855072463768115, | |
| "grad_norm": 0.8778862523158388, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7745, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 1.0273017034313925, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7723, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.47101449275362317, | |
| "grad_norm": 1.1202477015356398, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7689, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.5072463768115942, | |
| "grad_norm": 1.1947838008092264, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7588, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "grad_norm": 0.9801347289517877, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7641, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5797101449275363, | |
| "grad_norm": 0.6049825201268222, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7579, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.6159420289855072, | |
| "grad_norm": 0.7687560634971528, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7595, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 0.7372634515268928, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7536, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.6884057971014492, | |
| "grad_norm": 0.7700572406910111, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7531, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.7246376811594203, | |
| "grad_norm": 0.6355090382849566, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7521, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7608695652173914, | |
| "grad_norm": 0.6549502688413438, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7503, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.7971014492753623, | |
| "grad_norm": 0.582664220792735, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7551, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 0.6407186366739495, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7459, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 0.6331891386534093, | |
| "learning_rate": 5e-06, | |
| "loss": 0.744, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.9057971014492754, | |
| "grad_norm": 0.7299118855941761, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7374, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9420289855072463, | |
| "grad_norm": 0.567880243301988, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7458, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.9782608695652174, | |
| "grad_norm": 0.6907081447217029, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7358, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7438245415687561, | |
| "eval_runtime": 26.677, | |
| "eval_samples_per_second": 278.03, | |
| "eval_steps_per_second": 1.087, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 1.0144927536231885, | |
| "grad_norm": 0.8940503757064469, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7253, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.0507246376811594, | |
| "grad_norm": 0.7439488715557626, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6908, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 0.6352130108582714, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6918, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.1231884057971016, | |
| "grad_norm": 0.6093227815341429, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6951, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.1594202898550725, | |
| "grad_norm": 0.648343758224349, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6853, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.1956521739130435, | |
| "grad_norm": 0.7170626934482996, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6948, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.2318840579710144, | |
| "grad_norm": 0.669568534520701, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6845, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.2681159420289856, | |
| "grad_norm": 0.7720612077716418, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6873, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "grad_norm": 0.6321115893300935, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6889, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.3405797101449275, | |
| "grad_norm": 0.6385246244453152, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6886, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.3768115942028984, | |
| "grad_norm": 0.6322551870212395, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6902, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.4130434782608696, | |
| "grad_norm": 0.682130131174512, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6881, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.4492753623188406, | |
| "grad_norm": 0.6287023165159331, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6874, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.4855072463768115, | |
| "grad_norm": 0.6249521988426261, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6917, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "grad_norm": 0.6110574768057114, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6875, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.5579710144927537, | |
| "grad_norm": 0.771458385172311, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6905, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.5942028985507246, | |
| "grad_norm": 0.8776023781615215, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6934, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.6304347826086958, | |
| "grad_norm": 0.7459525790966258, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6927, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.579140354002018, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6857, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.7028985507246377, | |
| "grad_norm": 0.7077586289894291, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6914, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.5827410978723677, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6861, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.7753623188405796, | |
| "grad_norm": 0.697371946305589, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6894, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.8115942028985508, | |
| "grad_norm": 0.6086228815522553, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6905, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.8478260869565217, | |
| "grad_norm": 0.5925797270404988, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6875, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.8840579710144927, | |
| "grad_norm": 0.5774122986770631, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6864, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.9202898550724639, | |
| "grad_norm": 0.7752937398691316, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6848, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.9565217391304348, | |
| "grad_norm": 0.9039716311151348, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6851, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.9927536231884058, | |
| "grad_norm": 0.685112724319237, | |
| "learning_rate": 5e-06, | |
| "loss": 0.685, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7335925102233887, | |
| "eval_runtime": 26.8139, | |
| "eval_samples_per_second": 276.611, | |
| "eval_steps_per_second": 1.082, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 2.028985507246377, | |
| "grad_norm": 0.7665524028493372, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6385, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.0652173913043477, | |
| "grad_norm": 0.6043892824700962, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6304, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.101449275362319, | |
| "grad_norm": 0.7259867108096663, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6298, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.13768115942029, | |
| "grad_norm": 0.7510541142479693, | |
| "learning_rate": 5e-06, | |
| "loss": 0.635, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.1739130434782608, | |
| "grad_norm": 0.7187894788401495, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6341, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.210144927536232, | |
| "grad_norm": 0.6750049579183571, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6381, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.246376811594203, | |
| "grad_norm": 0.8490350516471592, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6324, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.282608695652174, | |
| "grad_norm": 0.7651747850436946, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6359, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.318840579710145, | |
| "grad_norm": 0.8525868267751531, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6327, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.355072463768116, | |
| "grad_norm": 0.9041979799840293, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6335, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.391304347826087, | |
| "grad_norm": 0.6631449903138956, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6333, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.427536231884058, | |
| "grad_norm": 0.6077513953951755, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6354, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.463768115942029, | |
| "grad_norm": 0.7123161501999767, | |
| "learning_rate": 5e-06, | |
| "loss": 0.635, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.7517439764623557, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6339, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.536231884057971, | |
| "grad_norm": 0.782865335447992, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6409, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.572463768115942, | |
| "grad_norm": 0.8659272969336265, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6364, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "grad_norm": 0.6169065680456968, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6322, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.644927536231884, | |
| "grad_norm": 0.637344472194744, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6394, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.681159420289855, | |
| "grad_norm": 0.723181305159317, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6423, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.717391304347826, | |
| "grad_norm": 0.6636365428175426, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6364, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.753623188405797, | |
| "grad_norm": 0.6833062497178735, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6393, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.789855072463768, | |
| "grad_norm": 0.6330454188928252, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6373, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.8260869565217392, | |
| "grad_norm": 0.6790151240239423, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6375, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.86231884057971, | |
| "grad_norm": 0.6977511814313268, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6356, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.898550724637681, | |
| "grad_norm": 0.5840071260671889, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6407, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.9347826086956523, | |
| "grad_norm": 0.6248935845380484, | |
| "learning_rate": 5e-06, | |
| "loss": 0.634, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.971014492753623, | |
| "grad_norm": 0.6181053130332258, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6412, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.7416301369667053, | |
| "eval_runtime": 26.232, | |
| "eval_samples_per_second": 282.746, | |
| "eval_steps_per_second": 1.106, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 828, | |
| "total_flos": 1386930839224320.0, | |
| "train_loss": 0.7053444857758601, | |
| "train_runtime": 5350.6579, | |
| "train_samples_per_second": 79.003, | |
| "train_steps_per_second": 0.155 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 828, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1386930839224320.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |