| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 38.0, | |
| "eval_steps": 500, | |
| "global_step": 20102, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1890359168241966, | |
| "grad_norm": 0.5107442140579224, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0482, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3780718336483932, | |
| "grad_norm": 0.5563424229621887, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8835, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5671077504725898, | |
| "grad_norm": 0.4735371172428131, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8512, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7561436672967864, | |
| "grad_norm": 0.4543495178222656, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8489, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.945179584120983, | |
| "grad_norm": 0.5007112622261047, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8439, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5733333333333334, | |
| "eval_loss": 1.6194567680358887, | |
| "eval_runtime": 5.4154, | |
| "eval_samples_per_second": 92.33, | |
| "eval_steps_per_second": 11.634, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 17.0, | |
| "eval_f1": 26.880952380952387, | |
| "step": 529 | |
| }, | |
| { | |
| "epoch": 1.1342155009451795, | |
| "grad_norm": 0.5181360244750977, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8217, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3232514177693762, | |
| "grad_norm": 1.1348918676376343, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7866, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5122873345935728, | |
| "grad_norm": 0.661758303642273, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7934, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7013232514177694, | |
| "grad_norm": 0.5634772777557373, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7731, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.8903591682419658, | |
| "grad_norm": 0.6698930263519287, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7969, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5751794871794872, | |
| "eval_loss": 1.603736162185669, | |
| "eval_runtime": 5.0986, | |
| "eval_samples_per_second": 98.066, | |
| "eval_steps_per_second": 12.356, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 17.6, | |
| "eval_f1": 27.218095238095238, | |
| "step": 1058 | |
| }, | |
| { | |
| "epoch": 2.0793950850661624, | |
| "grad_norm": 0.6265193223953247, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7416, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.268431001890359, | |
| "grad_norm": 0.812364935874939, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6957, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.4574669187145557, | |
| "grad_norm": 1.0972926616668701, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7033, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.6465028355387523, | |
| "grad_norm": 1.0831658840179443, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6935, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.835538752362949, | |
| "grad_norm": 1.0022494792938232, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6799, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.573025641025641, | |
| "eval_loss": 1.6428505182266235, | |
| "eval_runtime": 5.1488, | |
| "eval_samples_per_second": 97.111, | |
| "eval_steps_per_second": 12.236, | |
| "step": 1587 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 20.2, | |
| "eval_f1": 27.93809523809523, | |
| "step": 1587 | |
| }, | |
| { | |
| "epoch": 3.0245746691871456, | |
| "grad_norm": 0.8240587711334229, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6618, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.213610586011342, | |
| "grad_norm": 1.2994645833969116, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5736, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.402646502835539, | |
| "grad_norm": 1.0891159772872925, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5798, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.5916824196597354, | |
| "grad_norm": 1.3826475143432617, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5957, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.780718336483932, | |
| "grad_norm": 0.9901586174964905, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6026, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.9697542533081287, | |
| "grad_norm": 1.0607813596725464, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6106, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5708717948717948, | |
| "eval_loss": 1.7143596410751343, | |
| "eval_runtime": 5.3942, | |
| "eval_samples_per_second": 92.693, | |
| "eval_steps_per_second": 11.679, | |
| "step": 2116 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 19.0, | |
| "eval_f1": 27.71428571428571, | |
| "step": 2116 | |
| }, | |
| { | |
| "epoch": 4.158790170132325, | |
| "grad_norm": 1.3189178705215454, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5029, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.3478260869565215, | |
| "grad_norm": 1.062267541885376, | |
| "learning_rate": 5e-05, | |
| "loss": 1.51, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.536862003780718, | |
| "grad_norm": 1.3810843229293823, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4867, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.725897920604915, | |
| "grad_norm": 1.1991653442382812, | |
| "learning_rate": 5e-05, | |
| "loss": 1.501, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.914933837429111, | |
| "grad_norm": 1.310654878616333, | |
| "learning_rate": 5e-05, | |
| "loss": 1.5034, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5681025641025641, | |
| "eval_loss": 1.8328346014022827, | |
| "eval_runtime": 5.7763, | |
| "eval_samples_per_second": 86.561, | |
| "eval_steps_per_second": 10.907, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_exact_match": 18.8, | |
| "eval_f1": 26.6015873015873, | |
| "step": 2645 | |
| }, | |
| { | |
| "epoch": 5.103969754253308, | |
| "grad_norm": 1.1406631469726562, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4605, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.293005671077505, | |
| "grad_norm": 1.447221279144287, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4204, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.482041587901701, | |
| "grad_norm": 1.461799144744873, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4233, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.671077504725898, | |
| "grad_norm": 1.4184092283248901, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4265, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.8601134215500945, | |
| "grad_norm": 0.9850455522537231, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4139, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.564974358974359, | |
| "eval_loss": 1.931396484375, | |
| "eval_runtime": 5.3358, | |
| "eval_samples_per_second": 93.706, | |
| "eval_steps_per_second": 11.807, | |
| "step": 3174 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_exact_match": 17.6, | |
| "eval_f1": 25.464285714285715, | |
| "step": 3174 | |
| }, | |
| { | |
| "epoch": 6.049149338374291, | |
| "grad_norm": 1.2274116277694702, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4159, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.238185255198488, | |
| "grad_norm": 1.2107988595962524, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3579, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.427221172022684, | |
| "grad_norm": 1.2965952157974243, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3485, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.616257088846881, | |
| "grad_norm": 1.3035151958465576, | |
| "learning_rate": 5e-05, | |
| "loss": 1.36, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.805293005671078, | |
| "grad_norm": 1.1840296983718872, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3599, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.994328922495274, | |
| "grad_norm": 1.780287265777588, | |
| "learning_rate": 5e-05, | |
| "loss": 1.355, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.565076923076923, | |
| "eval_loss": 1.9659509658813477, | |
| "eval_runtime": 5.3127, | |
| "eval_samples_per_second": 94.114, | |
| "eval_steps_per_second": 11.858, | |
| "step": 3703 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_exact_match": 18.0, | |
| "eval_f1": 26.546666666666663, | |
| "step": 3703 | |
| }, | |
| { | |
| "epoch": 7.183364839319471, | |
| "grad_norm": 1.1608421802520752, | |
| "learning_rate": 5e-05, | |
| "loss": 1.298, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.3724007561436675, | |
| "grad_norm": 1.083270788192749, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3088, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.561436672967864, | |
| "grad_norm": 1.3489899635314941, | |
| "learning_rate": 5e-05, | |
| "loss": 1.299, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.750472589792061, | |
| "grad_norm": 1.1663447618484497, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2841, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.939508506616257, | |
| "grad_norm": 1.9308583736419678, | |
| "learning_rate": 5e-05, | |
| "loss": 1.314, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5616923076923077, | |
| "eval_loss": 2.1164443492889404, | |
| "eval_runtime": 5.2165, | |
| "eval_samples_per_second": 95.85, | |
| "eval_steps_per_second": 12.077, | |
| "step": 4232 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_exact_match": 17.0, | |
| "eval_f1": 24.659841269841277, | |
| "step": 4232 | |
| }, | |
| { | |
| "epoch": 8.128544423440454, | |
| "grad_norm": 1.109632134437561, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2558, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 8.31758034026465, | |
| "grad_norm": 1.1132491827011108, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2264, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.506616257088847, | |
| "grad_norm": 1.1017369031906128, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2649, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 1.0874963998794556, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2512, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.88468809073724, | |
| "grad_norm": 1.2617064714431763, | |
| "learning_rate": 5e-05, | |
| "loss": 1.261, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5622051282051282, | |
| "eval_loss": 2.1635327339172363, | |
| "eval_runtime": 5.0832, | |
| "eval_samples_per_second": 98.363, | |
| "eval_steps_per_second": 12.394, | |
| "step": 4761 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_exact_match": 17.8, | |
| "eval_f1": 25.451428571428576, | |
| "step": 4761 | |
| }, | |
| { | |
| "epoch": 9.073724007561436, | |
| "grad_norm": 1.2258027791976929, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2298, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.262759924385634, | |
| "grad_norm": 1.8261148929595947, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1748, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.45179584120983, | |
| "grad_norm": 1.1221072673797607, | |
| "learning_rate": 5e-05, | |
| "loss": 1.19, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.640831758034027, | |
| "grad_norm": 1.3320170640945435, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1932, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 9.829867674858223, | |
| "grad_norm": 1.194077730178833, | |
| "learning_rate": 5e-05, | |
| "loss": 1.217, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5612820512820513, | |
| "eval_loss": 2.244041919708252, | |
| "eval_runtime": 5.1606, | |
| "eval_samples_per_second": 96.889, | |
| "eval_steps_per_second": 12.208, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_exact_match": 18.4, | |
| "eval_f1": 26.067662337662348, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 10.01890359168242, | |
| "grad_norm": 1.1352814435958862, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2093, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 10.207939508506616, | |
| "grad_norm": 1.2741838693618774, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1467, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 10.396975425330814, | |
| "grad_norm": 1.6901249885559082, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1288, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.58601134215501, | |
| "grad_norm": 1.1944465637207031, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1399, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 10.775047258979207, | |
| "grad_norm": 1.2234572172164917, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1678, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 10.964083175803403, | |
| "grad_norm": 1.2781612873077393, | |
| "learning_rate": 5e-05, | |
| "loss": 1.146, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5618461538461539, | |
| "eval_loss": 2.296630859375, | |
| "eval_runtime": 5.1078, | |
| "eval_samples_per_second": 97.889, | |
| "eval_steps_per_second": 12.334, | |
| "step": 5819 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_exact_match": 16.8, | |
| "eval_f1": 23.63305916305917, | |
| "step": 5819 | |
| }, | |
| { | |
| "epoch": 11.1531190926276, | |
| "grad_norm": 1.138975739479065, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0842, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 11.342155009451796, | |
| "grad_norm": 1.727989912033081, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0806, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 11.531190926275993, | |
| "grad_norm": 1.3104358911514282, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0953, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 11.720226843100189, | |
| "grad_norm": 1.3629850149154663, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0945, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 11.909262759924385, | |
| "grad_norm": 1.3276056051254272, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1069, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.562, | |
| "eval_loss": 2.3118155002593994, | |
| "eval_runtime": 5.3837, | |
| "eval_samples_per_second": 92.873, | |
| "eval_steps_per_second": 11.702, | |
| "step": 6348 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_exact_match": 17.4, | |
| "eval_f1": 24.43924963924964, | |
| "step": 6348 | |
| }, | |
| { | |
| "epoch": 12.098298676748582, | |
| "grad_norm": 1.2608078718185425, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0418, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 12.287334593572778, | |
| "grad_norm": 1.4456945657730103, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0258, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 12.476370510396976, | |
| "grad_norm": 1.4195287227630615, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0255, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 12.665406427221171, | |
| "grad_norm": 1.2595598697662354, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0511, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 12.854442344045369, | |
| "grad_norm": 1.5901292562484741, | |
| "learning_rate": 5e-05, | |
| "loss": 1.045, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5614871794871795, | |
| "eval_loss": 2.3580987453460693, | |
| "eval_runtime": 5.3496, | |
| "eval_samples_per_second": 93.466, | |
| "eval_steps_per_second": 11.777, | |
| "step": 6877 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_exact_match": 17.2, | |
| "eval_f1": 24.350995670995676, | |
| "step": 6877 | |
| }, | |
| { | |
| "epoch": 13.043478260869565, | |
| "grad_norm": 1.471511721611023, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0404, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 13.232514177693762, | |
| "grad_norm": 1.2928757667541504, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9573, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.421550094517958, | |
| "grad_norm": 1.5545622110366821, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9583, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 13.610586011342155, | |
| "grad_norm": 1.6132512092590332, | |
| "learning_rate": 5e-05, | |
| "loss": 0.981, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 13.799621928166351, | |
| "grad_norm": 1.7548742294311523, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9999, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 13.988657844990549, | |
| "grad_norm": 1.4219907522201538, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0028, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5618974358974359, | |
| "eval_loss": 2.413728952407837, | |
| "eval_runtime": 5.0464, | |
| "eval_samples_per_second": 99.08, | |
| "eval_steps_per_second": 12.484, | |
| "step": 7406 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_exact_match": 16.6, | |
| "eval_f1": 24.201789321789327, | |
| "step": 7406 | |
| }, | |
| { | |
| "epoch": 14.177693761814744, | |
| "grad_norm": 1.5799965858459473, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9096, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.366729678638942, | |
| "grad_norm": 1.5961594581604004, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9075, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 14.555765595463138, | |
| "grad_norm": 1.5756512880325317, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9222, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 14.744801512287335, | |
| "grad_norm": 1.515251636505127, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9445, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 14.93383742911153, | |
| "grad_norm": 2.6804299354553223, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9153, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.5596923076923077, | |
| "eval_loss": 2.4756176471710205, | |
| "eval_runtime": 5.1956, | |
| "eval_samples_per_second": 96.236, | |
| "eval_steps_per_second": 12.126, | |
| "step": 7935 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_exact_match": 16.8, | |
| "eval_f1": 24.146709956709966, | |
| "step": 7935 | |
| }, | |
| { | |
| "epoch": 15.122873345935728, | |
| "grad_norm": 1.7786214351654053, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8874, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.311909262759924, | |
| "grad_norm": 1.709060788154602, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8538, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 15.500945179584122, | |
| "grad_norm": 2.3245434761047363, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8591, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 15.689981096408317, | |
| "grad_norm": 1.6101887226104736, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8593, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 15.879017013232515, | |
| "grad_norm": 1.5917792320251465, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8748, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5604615384615385, | |
| "eval_loss": 2.4434447288513184, | |
| "eval_runtime": 5.1072, | |
| "eval_samples_per_second": 97.901, | |
| "eval_steps_per_second": 12.336, | |
| "step": 8464 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_exact_match": 15.6, | |
| "eval_f1": 22.917027417027416, | |
| "step": 8464 | |
| }, | |
| { | |
| "epoch": 16.068052930056712, | |
| "grad_norm": 1.6529980897903442, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8413, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.257088846880908, | |
| "grad_norm": 1.931520938873291, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7863, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 16.446124763705104, | |
| "grad_norm": 1.8401894569396973, | |
| "learning_rate": 5e-05, | |
| "loss": 0.799, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 16.6351606805293, | |
| "grad_norm": 1.7952015399932861, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8238, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 16.8241965973535, | |
| "grad_norm": 1.8028634786605835, | |
| "learning_rate": 5e-05, | |
| "loss": 0.807, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5606153846153846, | |
| "eval_loss": 2.4520959854125977, | |
| "eval_runtime": 5.1769, | |
| "eval_samples_per_second": 96.584, | |
| "eval_steps_per_second": 12.17, | |
| "step": 8993 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_exact_match": 17.4, | |
| "eval_f1": 25.104011544011545, | |
| "step": 8993 | |
| }, | |
| { | |
| "epoch": 17.013232514177695, | |
| "grad_norm": 1.7197688817977905, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8249, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.20226843100189, | |
| "grad_norm": 1.768463134765625, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7304, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 1.9696024656295776, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7445, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 17.58034026465028, | |
| "grad_norm": 2.093703269958496, | |
| "learning_rate": 5e-05, | |
| "loss": 0.755, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 17.76937618147448, | |
| "grad_norm": 1.909550666809082, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7698, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 17.958412098298677, | |
| "grad_norm": 1.6835806369781494, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7711, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5603589743589743, | |
| "eval_loss": 2.473491668701172, | |
| "eval_runtime": 5.4257, | |
| "eval_samples_per_second": 92.155, | |
| "eval_steps_per_second": 11.611, | |
| "step": 9522 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_exact_match": 17.0, | |
| "eval_f1": 24.074487734487732, | |
| "step": 9522 | |
| }, | |
| { | |
| "epoch": 18.147448015122873, | |
| "grad_norm": 2.0483736991882324, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7048, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 18.33648393194707, | |
| "grad_norm": 1.987457275390625, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6863, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 18.525519848771268, | |
| "grad_norm": 2.1744563579559326, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7045, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 18.714555765595463, | |
| "grad_norm": 1.9871633052825928, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7076, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 18.90359168241966, | |
| "grad_norm": 1.8895001411437988, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7202, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.5601538461538461, | |
| "eval_loss": 2.5168700218200684, | |
| "eval_runtime": 5.0434, | |
| "eval_samples_per_second": 99.14, | |
| "eval_steps_per_second": 12.492, | |
| "step": 10051 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_exact_match": 17.2, | |
| "eval_f1": 24.260360750360757, | |
| "step": 10051 | |
| }, | |
| { | |
| "epoch": 19.092627599243855, | |
| "grad_norm": 1.7914209365844727, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6728, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 19.281663516068054, | |
| "grad_norm": 2.198495388031006, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6241, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 19.47069943289225, | |
| "grad_norm": 1.81365966796875, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6558, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 19.659735349716446, | |
| "grad_norm": 1.6571800708770752, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6652, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 19.84877126654064, | |
| "grad_norm": 1.6469954252243042, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6637, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5596410256410257, | |
| "eval_loss": 2.516073703765869, | |
| "eval_runtime": 5.8234, | |
| "eval_samples_per_second": 85.86, | |
| "eval_steps_per_second": 10.818, | |
| "step": 10580 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_exact_match": 17.6, | |
| "eval_f1": 25.225079365079367, | |
| "step": 10580 | |
| }, | |
| { | |
| "epoch": 20.03780718336484, | |
| "grad_norm": 1.7114077806472778, | |
| "learning_rate": 5e-05, | |
| "loss": 0.66, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 20.226843100189036, | |
| "grad_norm": 1.9962202310562134, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5938, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 20.415879017013232, | |
| "grad_norm": 2.397592067718506, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6042, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 20.604914933837428, | |
| "grad_norm": 2.0724213123321533, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6124, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 20.793950850661627, | |
| "grad_norm": 1.9385457038879395, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6078, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 20.982986767485823, | |
| "grad_norm": 2.034963607788086, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6257, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.5597948717948718, | |
| "eval_loss": 2.5497183799743652, | |
| "eval_runtime": 5.0442, | |
| "eval_samples_per_second": 99.123, | |
| "eval_steps_per_second": 12.489, | |
| "step": 11109 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_exact_match": 17.4, | |
| "eval_f1": 24.755238095238095, | |
| "step": 11109 | |
| }, | |
| { | |
| "epoch": 21.17202268431002, | |
| "grad_norm": 1.9768177270889282, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5458, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 21.361058601134214, | |
| "grad_norm": 2.14420747756958, | |
| "learning_rate": 5e-05, | |
| "loss": 0.558, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 21.550094517958414, | |
| "grad_norm": 2.5267951488494873, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5646, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 21.73913043478261, | |
| "grad_norm": 2.114386558532715, | |
| "learning_rate": 5e-05, | |
| "loss": 0.572, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 21.928166351606805, | |
| "grad_norm": 2.424217700958252, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5928, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.5593846153846154, | |
| "eval_loss": 2.579535722732544, | |
| "eval_runtime": 5.7849, | |
| "eval_samples_per_second": 86.433, | |
| "eval_steps_per_second": 10.89, | |
| "step": 11638 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_exact_match": 16.8, | |
| "eval_f1": 23.666507936507937, | |
| "step": 11638 | |
| }, | |
| { | |
| "epoch": 22.117202268431, | |
| "grad_norm": 1.7813427448272705, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5342, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 22.3062381852552, | |
| "grad_norm": 1.989479422569275, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5157, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 22.495274102079396, | |
| "grad_norm": 1.9694567918777466, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5236, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 22.68431001890359, | |
| "grad_norm": 2.0226142406463623, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5315, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 22.873345935727787, | |
| "grad_norm": 2.0204625129699707, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5378, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.558974358974359, | |
| "eval_loss": 2.5709009170532227, | |
| "eval_runtime": 5.0788, | |
| "eval_samples_per_second": 98.449, | |
| "eval_steps_per_second": 12.405, | |
| "step": 12167 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_exact_match": 16.4, | |
| "eval_f1": 23.69809523809524, | |
| "step": 12167 | |
| }, | |
| { | |
| "epoch": 23.062381852551987, | |
| "grad_norm": 2.0633811950683594, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5183, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 23.251417769376182, | |
| "grad_norm": 2.3566079139709473, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4761, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 23.440453686200378, | |
| "grad_norm": 2.368450880050659, | |
| "learning_rate": 5e-05, | |
| "loss": 0.487, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 23.629489603024574, | |
| "grad_norm": 2.1096372604370117, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4945, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 23.81852551984877, | |
| "grad_norm": 2.2136454582214355, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5035, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5586153846153846, | |
| "eval_loss": 2.5898056030273438, | |
| "eval_runtime": 5.7458, | |
| "eval_samples_per_second": 87.021, | |
| "eval_steps_per_second": 10.965, | |
| "step": 12696 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_exact_match": 17.8, | |
| "eval_f1": 24.89603174603175, | |
| "step": 12696 | |
| }, | |
| { | |
| "epoch": 24.00756143667297, | |
| "grad_norm": 2.317958116531372, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5106, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 24.196597353497165, | |
| "grad_norm": 2.0783536434173584, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4404, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 24.38563327032136, | |
| "grad_norm": 2.322791337966919, | |
| "learning_rate": 5e-05, | |
| "loss": 0.452, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 24.574669187145556, | |
| "grad_norm": 1.9763044118881226, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4617, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 24.763705103969755, | |
| "grad_norm": 2.462392568588257, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4731, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 24.95274102079395, | |
| "grad_norm": 2.2400307655334473, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4751, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.5594871794871795, | |
| "eval_loss": 2.611313581466675, | |
| "eval_runtime": 5.3564, | |
| "eval_samples_per_second": 93.346, | |
| "eval_steps_per_second": 11.762, | |
| "step": 13225 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_exact_match": 15.8, | |
| "eval_f1": 22.969047619047625, | |
| "step": 13225 | |
| }, | |
| { | |
| "epoch": 25.141776937618147, | |
| "grad_norm": 2.0931668281555176, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4299, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 25.330812854442343, | |
| "grad_norm": 2.7056140899658203, | |
| "learning_rate": 5e-05, | |
| "loss": 0.422, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 25.519848771266542, | |
| "grad_norm": 1.9907532930374146, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4374, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 25.708884688090738, | |
| "grad_norm": 2.3209221363067627, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4376, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 25.897920604914933, | |
| "grad_norm": 3.2339212894439697, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4432, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.5602051282051282, | |
| "eval_loss": 2.6182327270507812, | |
| "eval_runtime": 5.0321, | |
| "eval_samples_per_second": 99.362, | |
| "eval_steps_per_second": 12.52, | |
| "step": 13754 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_exact_match": 15.4, | |
| "eval_f1": 23.059047619047618, | |
| "step": 13754 | |
| }, | |
| { | |
| "epoch": 26.08695652173913, | |
| "grad_norm": 1.9935250282287598, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4265, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 26.27599243856333, | |
| "grad_norm": 2.5698649883270264, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3949, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 26.465028355387524, | |
| "grad_norm": 2.568392038345337, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4041, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 26.65406427221172, | |
| "grad_norm": 2.094651699066162, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4128, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 26.843100189035916, | |
| "grad_norm": 2.240649700164795, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4191, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.56, | |
| "eval_loss": 2.614851951599121, | |
| "eval_runtime": 5.636, | |
| "eval_samples_per_second": 88.715, | |
| "eval_steps_per_second": 11.178, | |
| "step": 14283 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_exact_match": 17.2, | |
| "eval_f1": 24.941948051948046, | |
| "step": 14283 | |
| }, | |
| { | |
| "epoch": 27.032136105860115, | |
| "grad_norm": 1.8800567388534546, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4135, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 27.22117202268431, | |
| "grad_norm": 2.525048017501831, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3717, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 27.410207939508506, | |
| "grad_norm": 2.3803513050079346, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3817, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 27.599243856332702, | |
| "grad_norm": 2.369356155395508, | |
| "learning_rate": 5e-05, | |
| "loss": 0.388, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 27.7882797731569, | |
| "grad_norm": 2.701702356338501, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3943, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 27.977315689981097, | |
| "grad_norm": 2.2755825519561768, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4015, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5592820512820513, | |
| "eval_loss": 2.653672218322754, | |
| "eval_runtime": 5.3543, | |
| "eval_samples_per_second": 93.383, | |
| "eval_steps_per_second": 11.766, | |
| "step": 14812 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_exact_match": 17.4, | |
| "eval_f1": 24.478888888888893, | |
| "step": 14812 | |
| }, | |
| { | |
| "epoch": 28.166351606805293, | |
| "grad_norm": 2.0458974838256836, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3566, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 28.35538752362949, | |
| "grad_norm": 2.040117025375366, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3602, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 28.544423440453688, | |
| "grad_norm": 1.947757363319397, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3666, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 28.733459357277884, | |
| "grad_norm": 2.0905520915985107, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3747, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 28.92249527410208, | |
| "grad_norm": 2.215851306915283, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3798, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.5600512820512821, | |
| "eval_loss": 2.6304192543029785, | |
| "eval_runtime": 5.0359, | |
| "eval_samples_per_second": 99.288, | |
| "eval_steps_per_second": 12.51, | |
| "step": 15341 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_exact_match": 16.2, | |
| "eval_f1": 23.909062049062054, | |
| "step": 15341 | |
| }, | |
| { | |
| "epoch": 29.111531190926275, | |
| "grad_norm": 2.5050926208496094, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3538, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 29.300567107750474, | |
| "grad_norm": 2.0059566497802734, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3365, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 29.48960302457467, | |
| "grad_norm": 2.7678000926971436, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3523, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 29.678638941398866, | |
| "grad_norm": 2.6234450340270996, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3552, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 29.86767485822306, | |
| "grad_norm": 1.9958972930908203, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3629, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.56, | |
| "eval_loss": 2.6483516693115234, | |
| "eval_runtime": 5.0621, | |
| "eval_samples_per_second": 98.774, | |
| "eval_steps_per_second": 12.446, | |
| "step": 15870 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_exact_match": 16.4, | |
| "eval_f1": 24.021746031746037, | |
| "step": 15870 | |
| }, | |
| { | |
| "epoch": 30.056710775047257, | |
| "grad_norm": 2.5568833351135254, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3505, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 30.245746691871457, | |
| "grad_norm": 2.0313165187835693, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3252, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 30.434782608695652, | |
| "grad_norm": 1.8152469396591187, | |
| "learning_rate": 5e-05, | |
| "loss": 0.335, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 30.623818525519848, | |
| "grad_norm": 2.0004210472106934, | |
| "learning_rate": 5e-05, | |
| "loss": 0.34, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 30.812854442344044, | |
| "grad_norm": 2.3558757305145264, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3474, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.5605641025641026, | |
| "eval_loss": 2.6599507331848145, | |
| "eval_runtime": 5.3592, | |
| "eval_samples_per_second": 93.297, | |
| "eval_steps_per_second": 11.755, | |
| "step": 16399 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_exact_match": 16.8, | |
| "eval_f1": 23.902857142857147, | |
| "step": 16399 | |
| }, | |
| { | |
| "epoch": 31.001890359168243, | |
| "grad_norm": 1.641440749168396, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3478, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 31.19092627599244, | |
| "grad_norm": 2.425560712814331, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3067, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 31.379962192816635, | |
| "grad_norm": 1.850221872329712, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3202, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 31.56899810964083, | |
| "grad_norm": 1.4244951009750366, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3301, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 31.75803402646503, | |
| "grad_norm": 2.093834638595581, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3293, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 31.947069943289225, | |
| "grad_norm": 2.0703086853027344, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3369, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5597435897435897, | |
| "eval_loss": 2.6675195693969727, | |
| "eval_runtime": 5.1201, | |
| "eval_samples_per_second": 97.655, | |
| "eval_steps_per_second": 12.305, | |
| "step": 16928 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_exact_match": 15.6, | |
| "eval_f1": 22.95126984126984, | |
| "step": 16928 | |
| }, | |
| { | |
| "epoch": 32.136105860113425, | |
| "grad_norm": 2.0121419429779053, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3134, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 32.32514177693762, | |
| "grad_norm": 1.8783822059631348, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3102, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 32.514177693761816, | |
| "grad_norm": 1.5319064855575562, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3089, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 32.70321361058601, | |
| "grad_norm": 2.6580235958099365, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3175, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 32.89224952741021, | |
| "grad_norm": 2.121335744857788, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3275, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.5603589743589743, | |
| "eval_loss": 2.658357858657837, | |
| "eval_runtime": 5.4806, | |
| "eval_samples_per_second": 91.231, | |
| "eval_steps_per_second": 11.495, | |
| "step": 17457 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_exact_match": 16.4, | |
| "eval_f1": 24.0452380952381, | |
| "step": 17457 | |
| }, | |
| { | |
| "epoch": 33.0812854442344, | |
| "grad_norm": 2.1282331943511963, | |
| "learning_rate": 5e-05, | |
| "loss": 0.311, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 33.2703213610586, | |
| "grad_norm": 2.212125062942505, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2951, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 33.459357277882795, | |
| "grad_norm": 3.389835834503174, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3068, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 33.648393194707, | |
| "grad_norm": 1.812578558921814, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3091, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 33.83742911153119, | |
| "grad_norm": 2.434966802597046, | |
| "learning_rate": 5e-05, | |
| "loss": 0.311, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.5600512820512821, | |
| "eval_loss": 2.676753520965576, | |
| "eval_runtime": 5.3579, | |
| "eval_samples_per_second": 93.32, | |
| "eval_steps_per_second": 11.758, | |
| "step": 17986 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_exact_match": 16.6, | |
| "eval_f1": 24.913650793650795, | |
| "step": 17986 | |
| }, | |
| { | |
| "epoch": 34.02646502835539, | |
| "grad_norm": 1.5352602005004883, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3142, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 34.215500945179585, | |
| "grad_norm": 1.4355374574661255, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2897, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 34.40453686200378, | |
| "grad_norm": 1.7993170022964478, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2923, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 34.593572778827976, | |
| "grad_norm": 1.9478981494903564, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3001, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "grad_norm": 3.045933246612549, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3034, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 34.97164461247637, | |
| "grad_norm": 1.8796017169952393, | |
| "learning_rate": 5e-05, | |
| "loss": 0.31, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.5601538461538461, | |
| "eval_loss": 2.6845180988311768, | |
| "eval_runtime": 5.7492, | |
| "eval_samples_per_second": 86.969, | |
| "eval_steps_per_second": 10.958, | |
| "step": 18515 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_exact_match": 16.2, | |
| "eval_f1": 23.91369408369409, | |
| "step": 18515 | |
| }, | |
| { | |
| "epoch": 35.16068052930057, | |
| "grad_norm": 1.7619277238845825, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2853, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 35.349716446124766, | |
| "grad_norm": 1.7933720350265503, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2853, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 35.53875236294896, | |
| "grad_norm": 1.476181149482727, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2955, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 35.72778827977316, | |
| "grad_norm": 1.6427425146102905, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2934, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 35.916824196597354, | |
| "grad_norm": 1.7125171422958374, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3009, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5604615384615385, | |
| "eval_loss": 2.6663565635681152, | |
| "eval_runtime": 5.0382, | |
| "eval_samples_per_second": 99.241, | |
| "eval_steps_per_second": 12.504, | |
| "step": 19044 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_exact_match": 16.0, | |
| "eval_f1": 23.221269841269848, | |
| "step": 19044 | |
| }, | |
| { | |
| "epoch": 36.10586011342155, | |
| "grad_norm": 1.757238507270813, | |
| "learning_rate": 5e-05, | |
| "loss": 0.286, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 36.294896030245745, | |
| "grad_norm": 1.9411671161651611, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2761, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 36.48393194706994, | |
| "grad_norm": 2.305509328842163, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2873, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 36.67296786389414, | |
| "grad_norm": 1.6525070667266846, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2881, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 36.86200378071834, | |
| "grad_norm": 1.562827706336975, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2959, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.5594871794871795, | |
| "eval_loss": 2.738389492034912, | |
| "eval_runtime": 5.0296, | |
| "eval_samples_per_second": 99.412, | |
| "eval_steps_per_second": 12.526, | |
| "step": 19573 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_exact_match": 16.0, | |
| "eval_f1": 23.01206349206349, | |
| "step": 19573 | |
| }, | |
| { | |
| "epoch": 37.051039697542535, | |
| "grad_norm": 2.588029146194458, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2885, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 37.24007561436673, | |
| "grad_norm": 2.2393579483032227, | |
| "learning_rate": 5e-05, | |
| "loss": 0.267, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 37.42911153119093, | |
| "grad_norm": 1.5349841117858887, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2787, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 37.61814744801512, | |
| "grad_norm": 1.8498040437698364, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2873, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 37.80718336483932, | |
| "grad_norm": 1.9959454536437988, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2879, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 37.996219281663514, | |
| "grad_norm": 1.814253330230713, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2927, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.5592820512820513, | |
| "eval_loss": 2.731750965118408, | |
| "eval_runtime": 5.032, | |
| "eval_samples_per_second": 99.364, | |
| "eval_steps_per_second": 12.52, | |
| "step": 20102 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_exact_match": 16.4, | |
| "eval_f1": 23.767460317460323, | |
| "step": 20102 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 26450, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2383494727280886e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |