| { | |
| "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": 2.3023250102996826, | |
| "learning_rate": 0.0002945454545454545, | |
| "loss": 0.6742, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004207868714496108, | |
| "eval_accuracy": 0.8339995741844177, | |
| "eval_loss": 0.5872370004653931, | |
| "eval_runtime": 559.2325, | |
| "eval_samples_per_second": 8.499, | |
| "eval_steps_per_second": 2.126, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "grad_norm": 10.144115447998047, | |
| "learning_rate": 0.00028909090909090904, | |
| "loss": 0.7741, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 1.0286434888839722, | |
| "eval_runtime": 548.7332, | |
| "eval_samples_per_second": 8.662, | |
| "eval_steps_per_second": 2.167, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "grad_norm": 4.225183486938477, | |
| "learning_rate": 0.0002836363636363636, | |
| "loss": 0.8739, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 0.6784296631813049, | |
| "eval_runtime": 552.4092, | |
| "eval_samples_per_second": 8.604, | |
| "eval_steps_per_second": 2.152, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "grad_norm": 2.9214963912963867, | |
| "learning_rate": 0.00027818181818181815, | |
| "loss": 0.7206, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 0.6782186627388, | |
| "eval_runtime": 553.9264, | |
| "eval_samples_per_second": 8.581, | |
| "eval_steps_per_second": 2.146, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "grad_norm": 4.568026542663574, | |
| "learning_rate": 0.0002727272727272727, | |
| "loss": 0.7165, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "eval_accuracy": 0.8114874958992004, | |
| "eval_loss": 0.5903932452201843, | |
| "eval_runtime": 556.1381, | |
| "eval_samples_per_second": 8.546, | |
| "eval_steps_per_second": 2.138, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "grad_norm": 1.1793783903121948, | |
| "learning_rate": 0.0002672727272727272, | |
| "loss": 0.6391, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "eval_accuracy": 0.5914159417152405, | |
| "eval_loss": 0.6900256276130676, | |
| "eval_runtime": 557.0755, | |
| "eval_samples_per_second": 8.532, | |
| "eval_steps_per_second": 2.134, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "grad_norm": 1.104693055152893, | |
| "learning_rate": 0.0002618181818181818, | |
| "loss": 0.6694, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "eval_accuracy": 0.8569324612617493, | |
| "eval_loss": 0.6000075340270996, | |
| "eval_runtime": 559.2951, | |
| "eval_samples_per_second": 8.498, | |
| "eval_steps_per_second": 2.126, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "grad_norm": 3.541372537612915, | |
| "learning_rate": 0.00025636363636363633, | |
| "loss": 0.5766, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "eval_accuracy": 0.8055964708328247, | |
| "eval_loss": 0.5426038503646851, | |
| "eval_runtime": 555.5844, | |
| "eval_samples_per_second": 8.555, | |
| "eval_steps_per_second": 2.14, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "grad_norm": 3.705471992492676, | |
| "learning_rate": 0.00025090909090909086, | |
| "loss": 0.4908, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "eval_accuracy": 0.8270565867424011, | |
| "eval_loss": 0.39803212881088257, | |
| "eval_runtime": 553.6423, | |
| "eval_samples_per_second": 8.585, | |
| "eval_steps_per_second": 2.148, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "grad_norm": 2.6912827491760254, | |
| "learning_rate": 0.00024545454545454545, | |
| "loss": 0.5775, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "eval_accuracy": 0.663791298866272, | |
| "eval_loss": 0.7920271158218384, | |
| "eval_runtime": 547.7028, | |
| "eval_samples_per_second": 8.678, | |
| "eval_steps_per_second": 2.171, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "grad_norm": 1.2508213520050049, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.5908, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "eval_accuracy": 0.861561119556427, | |
| "eval_loss": 0.3321113884449005, | |
| "eval_runtime": 548.2222, | |
| "eval_samples_per_second": 8.67, | |
| "eval_steps_per_second": 2.169, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "grad_norm": 3.2217397689819336, | |
| "learning_rate": 0.00023454545454545454, | |
| "loss": 0.3658, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "eval_accuracy": 0.8897538185119629, | |
| "eval_loss": 0.26357656717300415, | |
| "eval_runtime": 553.9916, | |
| "eval_samples_per_second": 8.58, | |
| "eval_steps_per_second": 2.146, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "grad_norm": 1.183059811592102, | |
| "learning_rate": 0.00022909090909090907, | |
| "loss": 0.4267, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "eval_accuracy": 0.8424153327941895, | |
| "eval_loss": 0.27429431676864624, | |
| "eval_runtime": 552.9034, | |
| "eval_samples_per_second": 8.596, | |
| "eval_steps_per_second": 2.15, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "grad_norm": 3.4732155799865723, | |
| "learning_rate": 0.00022363636363636363, | |
| "loss": 0.2191, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "eval_accuracy": 0.9164738059043884, | |
| "eval_loss": 0.28884753584861755, | |
| "eval_runtime": 552.6166, | |
| "eval_samples_per_second": 8.601, | |
| "eval_steps_per_second": 2.152, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "grad_norm": 2.1488490104675293, | |
| "learning_rate": 0.00021818181818181816, | |
| "loss": 0.25, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "eval_accuracy": 0.9124763011932373, | |
| "eval_loss": 0.36515381932258606, | |
| "eval_runtime": 551.2736, | |
| "eval_samples_per_second": 8.622, | |
| "eval_steps_per_second": 2.157, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "grad_norm": 2.5727055072784424, | |
| "learning_rate": 0.00021272727272727272, | |
| "loss": 0.4994, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "eval_accuracy": 0.908478856086731, | |
| "eval_loss": 0.24686028063297272, | |
| "eval_runtime": 555.0552, | |
| "eval_samples_per_second": 8.563, | |
| "eval_steps_per_second": 2.142, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "grad_norm": 2.7330338954925537, | |
| "learning_rate": 0.00020727272727272725, | |
| "loss": 0.274, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "eval_accuracy": 0.9383547306060791, | |
| "eval_loss": 0.16697467863559723, | |
| "eval_runtime": 551.1246, | |
| "eval_samples_per_second": 8.624, | |
| "eval_steps_per_second": 2.157, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "grad_norm": 21.431997299194336, | |
| "learning_rate": 0.0002018181818181818, | |
| "loss": 0.2721, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "eval_accuracy": 0.9185777306556702, | |
| "eval_loss": 0.2895517349243164, | |
| "eval_runtime": 550.1769, | |
| "eval_samples_per_second": 8.639, | |
| "eval_steps_per_second": 2.161, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "grad_norm": 6.493747711181641, | |
| "learning_rate": 0.00019636363636363634, | |
| "loss": 0.4545, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "eval_accuracy": 0.9242583513259888, | |
| "eval_loss": 0.32155150175094604, | |
| "eval_runtime": 551.3203, | |
| "eval_samples_per_second": 8.621, | |
| "eval_steps_per_second": 2.157, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "grad_norm": 3.465214490890503, | |
| "learning_rate": 0.0001909090909090909, | |
| "loss": 0.3221, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "eval_accuracy": 0.9358299970626831, | |
| "eval_loss": 0.20135502517223358, | |
| "eval_runtime": 551.1975, | |
| "eval_samples_per_second": 8.623, | |
| "eval_steps_per_second": 2.157, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "grad_norm": 4.727893352508545, | |
| "learning_rate": 0.00018545454545454543, | |
| "loss": 0.2348, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "eval_accuracy": 0.9309909343719482, | |
| "eval_loss": 0.2655259370803833, | |
| "eval_runtime": 546.4467, | |
| "eval_samples_per_second": 8.698, | |
| "eval_steps_per_second": 2.176, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "grad_norm": 2.3263726234436035, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 0.3247, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "eval_accuracy": 0.9417210221290588, | |
| "eval_loss": 0.1690194457769394, | |
| "eval_runtime": 546.9859, | |
| "eval_samples_per_second": 8.689, | |
| "eval_steps_per_second": 2.174, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "grad_norm": 0.32978275418281555, | |
| "learning_rate": 0.00017454545454545452, | |
| "loss": 0.1945, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "eval_accuracy": 0.897538423538208, | |
| "eval_loss": 0.23890452086925507, | |
| "eval_runtime": 556.5512, | |
| "eval_samples_per_second": 8.54, | |
| "eval_steps_per_second": 2.136, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "grad_norm": 0.2658408582210541, | |
| "learning_rate": 0.00016909090909090907, | |
| "loss": 0.2405, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "eval_accuracy": 0.9516094923019409, | |
| "eval_loss": 0.1562691628932953, | |
| "eval_runtime": 550.3268, | |
| "eval_samples_per_second": 8.637, | |
| "eval_steps_per_second": 2.161, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "grad_norm": 2.5755348205566406, | |
| "learning_rate": 0.0001636363636363636, | |
| "loss": 0.1715, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "eval_accuracy": 0.9362508058547974, | |
| "eval_loss": 0.2467575967311859, | |
| "eval_runtime": 552.8523, | |
| "eval_samples_per_second": 8.597, | |
| "eval_steps_per_second": 2.151, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "grad_norm": 4.770780563354492, | |
| "learning_rate": 0.00015818181818181816, | |
| "loss": 0.1866, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "eval_accuracy": 0.9654954671859741, | |
| "eval_loss": 0.16653123497962952, | |
| "eval_runtime": 553.6754, | |
| "eval_samples_per_second": 8.584, | |
| "eval_steps_per_second": 2.147, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "grad_norm": 0.04907585307955742, | |
| "learning_rate": 0.0001527272727272727, | |
| "loss": 0.2001, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "eval_accuracy": 0.9648643136024475, | |
| "eval_loss": 0.168188214302063, | |
| "eval_runtime": 558.4706, | |
| "eval_samples_per_second": 8.511, | |
| "eval_steps_per_second": 2.129, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "grad_norm": 0.042652346193790436, | |
| "learning_rate": 0.00014727272727272725, | |
| "loss": 0.1359, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "eval_accuracy": 0.9686513543128967, | |
| "eval_loss": 0.1295313686132431, | |
| "eval_runtime": 553.3401, | |
| "eval_samples_per_second": 8.59, | |
| "eval_steps_per_second": 2.149, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "grad_norm": 0.10347462445497513, | |
| "learning_rate": 0.0001418181818181818, | |
| "loss": 0.1633, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "eval_accuracy": 0.969072163105011, | |
| "eval_loss": 0.12678129971027374, | |
| "eval_runtime": 553.2115, | |
| "eval_samples_per_second": 8.592, | |
| "eval_steps_per_second": 2.149, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "grad_norm": 0.07808654755353928, | |
| "learning_rate": 0.00013636363636363634, | |
| "loss": 0.0169, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "eval_accuracy": 0.9709656834602356, | |
| "eval_loss": 0.1480005532503128, | |
| "eval_runtime": 556.5578, | |
| "eval_samples_per_second": 8.54, | |
| "eval_steps_per_second": 2.136, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "grad_norm": 5.326434135437012, | |
| "learning_rate": 0.0001309090909090909, | |
| "loss": 0.2247, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "eval_accuracy": 0.9625499844551086, | |
| "eval_loss": 0.1905258446931839, | |
| "eval_runtime": 561.2973, | |
| "eval_samples_per_second": 8.468, | |
| "eval_steps_per_second": 2.118, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "grad_norm": 0.11235077679157257, | |
| "learning_rate": 0.00012545454545454543, | |
| "loss": 0.0609, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "eval_accuracy": 0.9627603888511658, | |
| "eval_loss": 0.18458549678325653, | |
| "eval_runtime": 553.4853, | |
| "eval_samples_per_second": 8.587, | |
| "eval_steps_per_second": 2.148, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "grad_norm": 0.11561981588602066, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 0.0687, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "eval_accuracy": 0.9713864922523499, | |
| "eval_loss": 0.1371518075466156, | |
| "eval_runtime": 546.6765, | |
| "eval_samples_per_second": 8.694, | |
| "eval_steps_per_second": 2.175, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "grad_norm": 0.08527376502752304, | |
| "learning_rate": 0.00011454545454545453, | |
| "loss": 0.2002, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "eval_accuracy": 0.9711760878562927, | |
| "eval_loss": 0.13917790353298187, | |
| "eval_runtime": 557.0042, | |
| "eval_samples_per_second": 8.533, | |
| "eval_steps_per_second": 2.135, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "grad_norm": 0.09233067184686661, | |
| "learning_rate": 0.00010909090909090908, | |
| "loss": 0.0674, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "eval_accuracy": 0.9596044421195984, | |
| "eval_loss": 0.2038307785987854, | |
| "eval_runtime": 551.7111, | |
| "eval_samples_per_second": 8.615, | |
| "eval_steps_per_second": 2.155, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "grad_norm": 1.7652984857559204, | |
| "learning_rate": 0.00010363636363636362, | |
| "loss": 0.2449, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "eval_accuracy": 0.9699137210845947, | |
| "eval_loss": 0.1559370756149292, | |
| "eval_runtime": 551.9133, | |
| "eval_samples_per_second": 8.612, | |
| "eval_steps_per_second": 2.154, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "grad_norm": 0.8582727313041687, | |
| "learning_rate": 9.818181818181817e-05, | |
| "loss": 0.0794, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "eval_accuracy": 0.9697033166885376, | |
| "eval_loss": 0.10738001018762589, | |
| "eval_runtime": 550.561, | |
| "eval_samples_per_second": 8.633, | |
| "eval_steps_per_second": 2.16, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "grad_norm": 0.20773591101169586, | |
| "learning_rate": 9.272727272727271e-05, | |
| "loss": 0.3155, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "eval_accuracy": 0.9701241254806519, | |
| "eval_loss": 0.1280895173549652, | |
| "eval_runtime": 553.1498, | |
| "eval_samples_per_second": 8.593, | |
| "eval_steps_per_second": 2.15, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "grad_norm": 0.19470787048339844, | |
| "learning_rate": 8.727272727272726e-05, | |
| "loss": 0.0265, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "eval_accuracy": 0.9617083668708801, | |
| "eval_loss": 0.172711580991745, | |
| "eval_runtime": 555.6513, | |
| "eval_samples_per_second": 8.554, | |
| "eval_steps_per_second": 2.14, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "grad_norm": 2.273648262023926, | |
| "learning_rate": 8.18181818181818e-05, | |
| "loss": 0.1631, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "eval_accuracy": 0.9726488590240479, | |
| "eval_loss": 0.11321134120225906, | |
| "eval_runtime": 552.9935, | |
| "eval_samples_per_second": 8.595, | |
| "eval_steps_per_second": 2.15, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "grad_norm": 0.14013071358203888, | |
| "learning_rate": 7.636363636363635e-05, | |
| "loss": 0.0432, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "eval_accuracy": 0.9724384546279907, | |
| "eval_loss": 0.0995568335056305, | |
| "eval_runtime": 551.9628, | |
| "eval_samples_per_second": 8.611, | |
| "eval_steps_per_second": 2.154, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "grad_norm": 5.5448994636535645, | |
| "learning_rate": 7.09090909090909e-05, | |
| "loss": 0.0878, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "eval_accuracy": 0.9720176458358765, | |
| "eval_loss": 0.11234048753976822, | |
| "eval_runtime": 554.8087, | |
| "eval_samples_per_second": 8.567, | |
| "eval_steps_per_second": 2.143, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "grad_norm": 0.11822725087404251, | |
| "learning_rate": 6.545454545454545e-05, | |
| "loss": 0.1447, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "eval_accuracy": 0.9713864922523499, | |
| "eval_loss": 0.10238009691238403, | |
| "eval_runtime": 553.3152, | |
| "eval_samples_per_second": 8.59, | |
| "eval_steps_per_second": 2.149, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "grad_norm": 0.1717545986175537, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 0.1208, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "eval_accuracy": 0.971596896648407, | |
| "eval_loss": 0.09492182731628418, | |
| "eval_runtime": 550.8446, | |
| "eval_samples_per_second": 8.629, | |
| "eval_steps_per_second": 2.159, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "grad_norm": 0.7529041171073914, | |
| "learning_rate": 5.454545454545454e-05, | |
| "loss": 0.0643, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "eval_accuracy": 0.9720176458358765, | |
| "eval_loss": 0.08474315702915192, | |
| "eval_runtime": 558.3353, | |
| "eval_samples_per_second": 8.513, | |
| "eval_steps_per_second": 2.13, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "grad_norm": 0.20409902930259705, | |
| "learning_rate": 4.9090909090909084e-05, | |
| "loss": 0.0433, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "eval_accuracy": 0.9758047461509705, | |
| "eval_loss": 0.0837019681930542, | |
| "eval_runtime": 555.0163, | |
| "eval_samples_per_second": 8.564, | |
| "eval_steps_per_second": 2.142, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "grad_norm": 0.052959345281124115, | |
| "learning_rate": 4.363636363636363e-05, | |
| "loss": 0.1001, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "eval_accuracy": 0.9793814420700073, | |
| "eval_loss": 0.08561240136623383, | |
| "eval_runtime": 550.605, | |
| "eval_samples_per_second": 8.632, | |
| "eval_steps_per_second": 2.159, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "grad_norm": 0.19414858520030975, | |
| "learning_rate": 3.8181818181818174e-05, | |
| "loss": 0.0887, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "eval_accuracy": 0.9768567085266113, | |
| "eval_loss": 0.11111757904291153, | |
| "eval_runtime": 549.5968, | |
| "eval_samples_per_second": 8.648, | |
| "eval_steps_per_second": 2.163, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 0.03906171768903732, | |
| "learning_rate": 3.2727272727272725e-05, | |
| "loss": 0.0554, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "eval_accuracy": 0.9758047461509705, | |
| "eval_loss": 0.12151744961738586, | |
| "eval_runtime": 552.2439, | |
| "eval_samples_per_second": 8.607, | |
| "eval_steps_per_second": 2.153, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "grad_norm": 0.08566837012767792, | |
| "learning_rate": 2.727272727272727e-05, | |
| "loss": 0.0293, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "eval_accuracy": 0.975383996963501, | |
| "eval_loss": 0.12937308847904205, | |
| "eval_runtime": 558.1401, | |
| "eval_samples_per_second": 8.516, | |
| "eval_steps_per_second": 2.13, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "grad_norm": 0.09275167435407639, | |
| "learning_rate": 2.1818181818181814e-05, | |
| "loss": 0.0152, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "eval_accuracy": 0.9758047461509705, | |
| "eval_loss": 0.12702897191047668, | |
| "eval_runtime": 555.0185, | |
| "eval_samples_per_second": 8.564, | |
| "eval_steps_per_second": 2.142, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "grad_norm": 0.031154340133070946, | |
| "learning_rate": 1.6363636363636363e-05, | |
| "loss": 0.1286, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "eval_accuracy": 0.9747527837753296, | |
| "eval_loss": 0.133098304271698, | |
| "eval_runtime": 554.216, | |
| "eval_samples_per_second": 8.576, | |
| "eval_steps_per_second": 2.145, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "grad_norm": 0.031359899789094925, | |
| "learning_rate": 1.0909090909090907e-05, | |
| "loss": 0.0186, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "eval_accuracy": 0.9737008213996887, | |
| "eval_loss": 0.1388082504272461, | |
| "eval_runtime": 549.5036, | |
| "eval_samples_per_second": 8.65, | |
| "eval_steps_per_second": 2.164, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "grad_norm": 0.03772009164094925, | |
| "learning_rate": 5.454545454545454e-06, | |
| "loss": 0.0299, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "eval_accuracy": 0.9732800126075745, | |
| "eval_loss": 0.14506614208221436, | |
| "eval_runtime": 554.9274, | |
| "eval_samples_per_second": 8.565, | |
| "eval_steps_per_second": 2.143, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "grad_norm": 0.07556959241628647, | |
| "learning_rate": 0.0, | |
| "loss": 0.0536, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "eval_accuracy": 0.972859263420105, | |
| "eval_loss": 0.14720246195793152, | |
| "eval_runtime": 548.0681, | |
| "eval_samples_per_second": 8.672, | |
| "eval_steps_per_second": 2.169, | |
| "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 | |
| } | |