| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23143277929728592, | |
| "eval_steps": 10, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004207868714496108, | |
| "grad_norm": 3.7386245727539062, | |
| "learning_rate": 0.00019636363636363636, | |
| "loss": 0.6798, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004207868714496108, | |
| "eval_accuracy": 0.6183463335037231, | |
| "eval_loss": 0.6594762802124023, | |
| "eval_runtime": 597.7321, | |
| "eval_samples_per_second": 7.952, | |
| "eval_steps_per_second": 1.989, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "grad_norm": 23.72800636291504, | |
| "learning_rate": 0.00019272727272727274, | |
| "loss": 0.734, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "eval_accuracy": 0.6175047159194946, | |
| "eval_loss": 0.6504878401756287, | |
| "eval_runtime": 581.0465, | |
| "eval_samples_per_second": 8.18, | |
| "eval_steps_per_second": 2.046, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "grad_norm": 9.25478458404541, | |
| "learning_rate": 0.0001890909090909091, | |
| "loss": 0.6702, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "eval_accuracy": 0.6919839978218079, | |
| "eval_loss": 0.5777503252029419, | |
| "eval_runtime": 578.0228, | |
| "eval_samples_per_second": 8.223, | |
| "eval_steps_per_second": 2.057, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "grad_norm": 7.304983615875244, | |
| "learning_rate": 0.00018545454545454545, | |
| "loss": 0.4556, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "eval_accuracy": 0.8396801948547363, | |
| "eval_loss": 0.40889662504196167, | |
| "eval_runtime": 571.8895, | |
| "eval_samples_per_second": 8.311, | |
| "eval_steps_per_second": 2.079, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "grad_norm": 7.751391887664795, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.569, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "eval_accuracy": 0.7944456338882446, | |
| "eval_loss": 0.4470515251159668, | |
| "eval_runtime": 573.7399, | |
| "eval_samples_per_second": 8.284, | |
| "eval_steps_per_second": 2.072, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "grad_norm": 1.6625334024429321, | |
| "learning_rate": 0.0001781818181818182, | |
| "loss": 0.4117, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "eval_accuracy": 0.8788133859634399, | |
| "eval_loss": 0.2980070412158966, | |
| "eval_runtime": 568.2097, | |
| "eval_samples_per_second": 8.365, | |
| "eval_steps_per_second": 2.093, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "grad_norm": 3.0882933139801025, | |
| "learning_rate": 0.00017454545454545454, | |
| "loss": 0.4465, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "eval_accuracy": 0.8990111351013184, | |
| "eval_loss": 0.24757108092308044, | |
| "eval_runtime": 561.4024, | |
| "eval_samples_per_second": 8.466, | |
| "eval_steps_per_second": 2.118, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "grad_norm": 1.2464115619659424, | |
| "learning_rate": 0.0001709090909090909, | |
| "loss": 0.4248, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "eval_accuracy": 0.9202609062194824, | |
| "eval_loss": 0.2152455598115921, | |
| "eval_runtime": 561.7534, | |
| "eval_samples_per_second": 8.461, | |
| "eval_steps_per_second": 2.117, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "grad_norm": 3.7030742168426514, | |
| "learning_rate": 0.00016727272727272728, | |
| "loss": 0.2108, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "eval_accuracy": 0.9377235174179077, | |
| "eval_loss": 0.18127086758613586, | |
| "eval_runtime": 560.5533, | |
| "eval_samples_per_second": 8.479, | |
| "eval_steps_per_second": 2.121, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "grad_norm": 3.4107439517974854, | |
| "learning_rate": 0.00016363636363636366, | |
| "loss": 0.3354, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "eval_accuracy": 0.9215232729911804, | |
| "eval_loss": 0.19136299192905426, | |
| "eval_runtime": 554.2596, | |
| "eval_samples_per_second": 8.575, | |
| "eval_steps_per_second": 2.145, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "grad_norm": 9.938206672668457, | |
| "learning_rate": 0.00016, | |
| "loss": 0.2655, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "eval_accuracy": 0.9354091882705688, | |
| "eval_loss": 0.1784314066171646, | |
| "eval_runtime": 556.5406, | |
| "eval_samples_per_second": 8.54, | |
| "eval_steps_per_second": 2.136, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "grad_norm": 7.809841156005859, | |
| "learning_rate": 0.00015636363636363637, | |
| "loss": 0.3885, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "eval_accuracy": 0.8979591727256775, | |
| "eval_loss": 0.2238394021987915, | |
| "eval_runtime": 558.4451, | |
| "eval_samples_per_second": 8.511, | |
| "eval_steps_per_second": 2.129, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "grad_norm": 0.18238992989063263, | |
| "learning_rate": 0.00015272727272727275, | |
| "loss": 0.1268, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "eval_accuracy": 0.965285062789917, | |
| "eval_loss": 0.08994818478822708, | |
| "eval_runtime": 564.2804, | |
| "eval_samples_per_second": 8.423, | |
| "eval_steps_per_second": 2.107, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "grad_norm": 21.266611099243164, | |
| "learning_rate": 0.0001490909090909091, | |
| "loss": 0.222, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "eval_accuracy": 0.9591836929321289, | |
| "eval_loss": 0.2074023336172104, | |
| "eval_runtime": 563.4722, | |
| "eval_samples_per_second": 8.435, | |
| "eval_steps_per_second": 2.11, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "grad_norm": 11.509130477905273, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 0.4098, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "eval_accuracy": 0.8276877999305725, | |
| "eval_loss": 0.7424523234367371, | |
| "eval_runtime": 560.1213, | |
| "eval_samples_per_second": 8.486, | |
| "eval_steps_per_second": 2.123, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "grad_norm": 7.520476818084717, | |
| "learning_rate": 0.00014181818181818184, | |
| "loss": 0.4331, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "eval_accuracy": 0.9326741099357605, | |
| "eval_loss": 0.2065199315547943, | |
| "eval_runtime": 562.1855, | |
| "eval_samples_per_second": 8.455, | |
| "eval_steps_per_second": 2.115, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "grad_norm": 12.3318510055542, | |
| "learning_rate": 0.0001381818181818182, | |
| "loss": 0.6336, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "eval_accuracy": 0.859036386013031, | |
| "eval_loss": 0.29903730750083923, | |
| "eval_runtime": 564.8753, | |
| "eval_samples_per_second": 8.414, | |
| "eval_steps_per_second": 2.105, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "grad_norm": 7.766119956970215, | |
| "learning_rate": 0.00013454545454545455, | |
| "loss": 0.2881, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "eval_accuracy": 0.948032796382904, | |
| "eval_loss": 0.11360938847064972, | |
| "eval_runtime": 565.4404, | |
| "eval_samples_per_second": 8.406, | |
| "eval_steps_per_second": 2.103, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "grad_norm": 6.830145359039307, | |
| "learning_rate": 0.00013090909090909093, | |
| "loss": 0.277, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "eval_accuracy": 0.9400378465652466, | |
| "eval_loss": 0.209593266248703, | |
| "eval_runtime": 562.6503, | |
| "eval_samples_per_second": 8.448, | |
| "eval_steps_per_second": 2.113, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "grad_norm": 4.35469913482666, | |
| "learning_rate": 0.00012727272727272728, | |
| "loss": 0.1586, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "eval_accuracy": 0.9337260723114014, | |
| "eval_loss": 0.23578013479709625, | |
| "eval_runtime": 562.1129, | |
| "eval_samples_per_second": 8.456, | |
| "eval_steps_per_second": 2.115, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "grad_norm": 2.565586566925049, | |
| "learning_rate": 0.00012363636363636364, | |
| "loss": 0.1664, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "eval_accuracy": 0.967809796333313, | |
| "eval_loss": 0.11096801608800888, | |
| "eval_runtime": 556.8442, | |
| "eval_samples_per_second": 8.536, | |
| "eval_steps_per_second": 2.135, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "grad_norm": 12.48508358001709, | |
| "learning_rate": 0.00012, | |
| "loss": 0.3419, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "eval_accuracy": 0.9421418309211731, | |
| "eval_loss": 0.14782342314720154, | |
| "eval_runtime": 560.1835, | |
| "eval_samples_per_second": 8.485, | |
| "eval_steps_per_second": 2.123, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "grad_norm": 10.233806610107422, | |
| "learning_rate": 0.00011636363636363636, | |
| "loss": 0.187, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "eval_accuracy": 0.9579213261604309, | |
| "eval_loss": 0.11080725491046906, | |
| "eval_runtime": 563.1087, | |
| "eval_samples_per_second": 8.441, | |
| "eval_steps_per_second": 2.111, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "grad_norm": 0.16775061190128326, | |
| "learning_rate": 0.00011272727272727272, | |
| "loss": 0.1834, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "eval_accuracy": 0.8727119565010071, | |
| "eval_loss": 0.44505125284194946, | |
| "eval_runtime": 564.6564, | |
| "eval_samples_per_second": 8.418, | |
| "eval_steps_per_second": 2.106, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "grad_norm": 5.313834190368652, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 0.2383, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "eval_accuracy": 0.9726488590240479, | |
| "eval_loss": 0.07795428484678268, | |
| "eval_runtime": 565.2854, | |
| "eval_samples_per_second": 8.408, | |
| "eval_steps_per_second": 2.103, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "grad_norm": 0.07698975503444672, | |
| "learning_rate": 0.00010545454545454545, | |
| "loss": 0.2157, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "eval_accuracy": 0.9758047461509705, | |
| "eval_loss": 0.0759287104010582, | |
| "eval_runtime": 564.058, | |
| "eval_samples_per_second": 8.426, | |
| "eval_steps_per_second": 2.108, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "grad_norm": 0.13711199164390564, | |
| "learning_rate": 0.00010181818181818181, | |
| "loss": 0.2665, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "eval_accuracy": 0.9787502884864807, | |
| "eval_loss": 0.06089087575674057, | |
| "eval_runtime": 563.7962, | |
| "eval_samples_per_second": 8.43, | |
| "eval_steps_per_second": 2.109, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "grad_norm": 0.17997244000434875, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.1226, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "eval_accuracy": 0.9808542132377625, | |
| "eval_loss": 0.052085429430007935, | |
| "eval_runtime": 564.5871, | |
| "eval_samples_per_second": 8.419, | |
| "eval_steps_per_second": 2.106, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "grad_norm": 0.1597593128681183, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.1287, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "eval_accuracy": 0.9558174014091492, | |
| "eval_loss": 0.09927653521299362, | |
| "eval_runtime": 563.2882, | |
| "eval_samples_per_second": 8.438, | |
| "eval_steps_per_second": 2.111, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "grad_norm": 14.694509506225586, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.1088, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "eval_accuracy": 0.9309909343719482, | |
| "eval_loss": 0.19473901391029358, | |
| "eval_runtime": 565.488, | |
| "eval_samples_per_second": 8.405, | |
| "eval_steps_per_second": 2.103, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "grad_norm": 4.671857833862305, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.1778, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "eval_accuracy": 0.9814853668212891, | |
| "eval_loss": 0.057908281683921814, | |
| "eval_runtime": 564.4128, | |
| "eval_samples_per_second": 8.421, | |
| "eval_steps_per_second": 2.107, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "grad_norm": 1.1494044065475464, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.1686, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "eval_accuracy": 0.9798022508621216, | |
| "eval_loss": 0.06721032410860062, | |
| "eval_runtime": 563.963, | |
| "eval_samples_per_second": 8.428, | |
| "eval_steps_per_second": 2.108, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "grad_norm": 20.857423782348633, | |
| "learning_rate": 8e-05, | |
| "loss": 0.19, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "eval_accuracy": 0.9793814420700073, | |
| "eval_loss": 0.07669834792613983, | |
| "eval_runtime": 558.0764, | |
| "eval_samples_per_second": 8.517, | |
| "eval_steps_per_second": 2.131, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "grad_norm": 0.02689620479941368, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.0889, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "eval_accuracy": 0.9776982665061951, | |
| "eval_loss": 0.0889260396361351, | |
| "eval_runtime": 561.1157, | |
| "eval_samples_per_second": 8.471, | |
| "eval_steps_per_second": 2.119, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "grad_norm": 0.01651788130402565, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.0046, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "eval_accuracy": 0.9774879217147827, | |
| "eval_loss": 0.0977482944726944, | |
| "eval_runtime": 557.4467, | |
| "eval_samples_per_second": 8.526, | |
| "eval_steps_per_second": 2.133, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "grad_norm": 0.49462929368019104, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.2327, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "eval_accuracy": 0.9833789467811584, | |
| "eval_loss": 0.05215293914079666, | |
| "eval_runtime": 559.0767, | |
| "eval_samples_per_second": 8.502, | |
| "eval_steps_per_second": 2.127, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "grad_norm": 1.9083374738693237, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.021, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "eval_accuracy": 0.9629707336425781, | |
| "eval_loss": 0.11694369465112686, | |
| "eval_runtime": 559.3448, | |
| "eval_samples_per_second": 8.497, | |
| "eval_steps_per_second": 2.126, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "grad_norm": 23.373567581176758, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.2001, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "eval_accuracy": 0.9829581379890442, | |
| "eval_loss": 0.060037508606910706, | |
| "eval_runtime": 560.337, | |
| "eval_samples_per_second": 8.482, | |
| "eval_steps_per_second": 2.122, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "grad_norm": 0.10927582532167435, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.0711, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "eval_accuracy": 0.9859036207199097, | |
| "eval_loss": 0.04903951659798622, | |
| "eval_runtime": 555.1166, | |
| "eval_samples_per_second": 8.562, | |
| "eval_steps_per_second": 2.142, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "grad_norm": 1.898078203201294, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.0983, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "eval_accuracy": 0.9831685423851013, | |
| "eval_loss": 0.045510243624448776, | |
| "eval_runtime": 558.1218, | |
| "eval_samples_per_second": 8.516, | |
| "eval_steps_per_second": 2.13, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "grad_norm": 0.12654219567775726, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.0548, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "eval_accuracy": 0.9888491630554199, | |
| "eval_loss": 0.04250233247876167, | |
| "eval_runtime": 558.6685, | |
| "eval_samples_per_second": 8.508, | |
| "eval_steps_per_second": 2.128, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "grad_norm": 2.7145519256591797, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.0505, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "eval_accuracy": 0.9880075454711914, | |
| "eval_loss": 0.04300825670361519, | |
| "eval_runtime": 559.6635, | |
| "eval_samples_per_second": 8.493, | |
| "eval_steps_per_second": 2.124, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "grad_norm": 0.08013765513896942, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.2609, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "eval_accuracy": 0.9899011254310608, | |
| "eval_loss": 0.03251320868730545, | |
| "eval_runtime": 560.0268, | |
| "eval_samples_per_second": 8.487, | |
| "eval_steps_per_second": 2.123, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "grad_norm": 0.1592099815607071, | |
| "learning_rate": 4e-05, | |
| "loss": 0.058, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "eval_accuracy": 0.9783294796943665, | |
| "eval_loss": 0.0525953434407711, | |
| "eval_runtime": 553.5191, | |
| "eval_samples_per_second": 8.587, | |
| "eval_steps_per_second": 2.148, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "grad_norm": 0.064529187977314, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.1082, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "eval_accuracy": 0.9888491630554199, | |
| "eval_loss": 0.03384377807378769, | |
| "eval_runtime": 557.5032, | |
| "eval_samples_per_second": 8.526, | |
| "eval_steps_per_second": 2.133, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "grad_norm": 0.35138818621635437, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.03, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "eval_accuracy": 0.9924258589744568, | |
| "eval_loss": 0.029149439185857773, | |
| "eval_runtime": 559.8975, | |
| "eval_samples_per_second": 8.489, | |
| "eval_steps_per_second": 2.124, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "grad_norm": 0.020687425509095192, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.0702, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "eval_accuracy": 0.9930570125579834, | |
| "eval_loss": 0.027508797124028206, | |
| "eval_runtime": 558.5715, | |
| "eval_samples_per_second": 8.509, | |
| "eval_steps_per_second": 2.129, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "grad_norm": 0.08153369277715683, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.0412, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "eval_accuracy": 0.9924258589744568, | |
| "eval_loss": 0.02858472615480423, | |
| "eval_runtime": 557.0232, | |
| "eval_samples_per_second": 8.533, | |
| "eval_steps_per_second": 2.135, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 0.027780650183558464, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.0056, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "eval_accuracy": 0.9922154545783997, | |
| "eval_loss": 0.028572723269462585, | |
| "eval_runtime": 557.9364, | |
| "eval_samples_per_second": 8.519, | |
| "eval_steps_per_second": 2.131, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "grad_norm": 0.013943745754659176, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0202, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "eval_accuracy": 0.9934778213500977, | |
| "eval_loss": 0.02832806296646595, | |
| "eval_runtime": 590.2498, | |
| "eval_samples_per_second": 8.053, | |
| "eval_steps_per_second": 2.014, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "grad_norm": 0.06914212554693222, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.0032, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "eval_accuracy": 0.9932674169540405, | |
| "eval_loss": 0.02869958057999611, | |
| "eval_runtime": 574.9638, | |
| "eval_samples_per_second": 8.267, | |
| "eval_steps_per_second": 2.068, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "grad_norm": 0.014089811593294144, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.0932, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "eval_accuracy": 0.9924258589744568, | |
| "eval_loss": 0.029511602595448494, | |
| "eval_runtime": 574.493, | |
| "eval_samples_per_second": 8.273, | |
| "eval_steps_per_second": 2.07, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "grad_norm": 0.006211976986378431, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.0597, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "eval_accuracy": 0.9926362037658691, | |
| "eval_loss": 0.029444053769111633, | |
| "eval_runtime": 566.9552, | |
| "eval_samples_per_second": 8.383, | |
| "eval_steps_per_second": 2.097, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "grad_norm": 0.030956320464611053, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.0069, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "eval_accuracy": 0.9926362037658691, | |
| "eval_loss": 0.02989383228123188, | |
| "eval_runtime": 560.5938, | |
| "eval_samples_per_second": 8.479, | |
| "eval_steps_per_second": 2.121, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "grad_norm": 0.10614718496799469, | |
| "learning_rate": 0.0, | |
| "loss": 0.024, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "eval_accuracy": 0.9924258589744568, | |
| "eval_loss": 0.030369577929377556, | |
| "eval_runtime": 551.1375, | |
| "eval_samples_per_second": 8.624, | |
| "eval_steps_per_second": 2.157, | |
| "step": 550 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.208452205931052e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |