| { | |
| "best_metric": 1.022267460823059, | |
| "best_model_checkpoint": "./model_fine-tune/glot/xlm-r/ben-Beng/checkpoint-98500", | |
| "epoch": 29.60625187856928, | |
| "eval_steps": 500, | |
| "global_step": 98500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15028554253080853, | |
| "grad_norm": 4.279157638549805, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.5206, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15028554253080853, | |
| "eval_accuracy": 0.7123080245508447, | |
| "eval_loss": 1.6415441036224365, | |
| "eval_runtime": 108.0973, | |
| "eval_samples_per_second": 220.635, | |
| "eval_steps_per_second": 6.901, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.30057108506161706, | |
| "grad_norm": 3.948293685913086, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.4505, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.30057108506161706, | |
| "eval_accuracy": 0.7195765918792126, | |
| "eval_loss": 1.6040955781936646, | |
| "eval_runtime": 108.1918, | |
| "eval_samples_per_second": 220.442, | |
| "eval_steps_per_second": 6.895, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4508566275924256, | |
| "grad_norm": 3.838204860687256, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.4092, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.4508566275924256, | |
| "eval_accuracy": 0.7255862618332372, | |
| "eval_loss": 1.5711835622787476, | |
| "eval_runtime": 105.314, | |
| "eval_samples_per_second": 226.466, | |
| "eval_steps_per_second": 7.084, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6011421701232341, | |
| "grad_norm": 3.7270383834838867, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.3738, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6011421701232341, | |
| "eval_accuracy": 0.7298732586939131, | |
| "eval_loss": 1.562820315361023, | |
| "eval_runtime": 104.3524, | |
| "eval_samples_per_second": 228.552, | |
| "eval_steps_per_second": 7.149, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7514277126540427, | |
| "grad_norm": 3.720855474472046, | |
| "learning_rate": 9.75e-05, | |
| "loss": 1.3447, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7514277126540427, | |
| "eval_accuracy": 0.7334076213509759, | |
| "eval_loss": 1.533949613571167, | |
| "eval_runtime": 105.6055, | |
| "eval_samples_per_second": 225.84, | |
| "eval_steps_per_second": 7.064, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.9017132551848512, | |
| "grad_norm": 3.9192264080047607, | |
| "learning_rate": 9.7e-05, | |
| "loss": 1.3215, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9017132551848512, | |
| "eval_accuracy": 0.737443643353561, | |
| "eval_loss": 1.5036447048187256, | |
| "eval_runtime": 103.4988, | |
| "eval_samples_per_second": 230.437, | |
| "eval_steps_per_second": 7.208, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0519987977156597, | |
| "grad_norm": 3.8544235229492188, | |
| "learning_rate": 9.65e-05, | |
| "loss": 1.3074, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.0519987977156597, | |
| "eval_accuracy": 0.7401312504576956, | |
| "eval_loss": 1.4803341627120972, | |
| "eval_runtime": 104.2256, | |
| "eval_samples_per_second": 228.831, | |
| "eval_steps_per_second": 7.158, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.2022843402464682, | |
| "grad_norm": 3.2664737701416016, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.2826, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.2022843402464682, | |
| "eval_accuracy": 0.7433592201310665, | |
| "eval_loss": 1.475807785987854, | |
| "eval_runtime": 104.5255, | |
| "eval_samples_per_second": 228.174, | |
| "eval_steps_per_second": 7.137, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.3525698827772767, | |
| "grad_norm": 3.448021650314331, | |
| "learning_rate": 9.55e-05, | |
| "loss": 1.2676, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.3525698827772767, | |
| "eval_accuracy": 0.7454010087654052, | |
| "eval_loss": 1.467736005783081, | |
| "eval_runtime": 106.1685, | |
| "eval_samples_per_second": 224.643, | |
| "eval_steps_per_second": 7.027, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.5028554253080855, | |
| "grad_norm": 3.6344072818756104, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.2462, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.5028554253080855, | |
| "eval_accuracy": 0.7480509601003421, | |
| "eval_loss": 1.454475998878479, | |
| "eval_runtime": 107.2463, | |
| "eval_samples_per_second": 222.385, | |
| "eval_steps_per_second": 6.956, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.653140967838894, | |
| "grad_norm": 3.191260576248169, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 1.2353, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.653140967838894, | |
| "eval_accuracy": 0.7502682619699697, | |
| "eval_loss": 1.4226497411727905, | |
| "eval_runtime": 107.1587, | |
| "eval_samples_per_second": 222.567, | |
| "eval_steps_per_second": 6.962, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.8034265103697025, | |
| "grad_norm": 3.074147939682007, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.2227, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.8034265103697025, | |
| "eval_accuracy": 0.7523797476360646, | |
| "eval_loss": 1.4185426235198975, | |
| "eval_runtime": 107.8236, | |
| "eval_samples_per_second": 221.195, | |
| "eval_steps_per_second": 6.919, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.953712052900511, | |
| "grad_norm": 3.0745174884796143, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 1.2179, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.953712052900511, | |
| "eval_accuracy": 0.7525899025066154, | |
| "eval_loss": 1.4250199794769287, | |
| "eval_runtime": 103.869, | |
| "eval_samples_per_second": 229.616, | |
| "eval_steps_per_second": 7.182, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.1039975954313195, | |
| "grad_norm": 3.809814453125, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 1.1973, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.1039975954313195, | |
| "eval_accuracy": 0.7544079430376059, | |
| "eval_loss": 1.4045050144195557, | |
| "eval_runtime": 103.9514, | |
| "eval_samples_per_second": 229.434, | |
| "eval_steps_per_second": 7.176, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.254283137962128, | |
| "grad_norm": 3.262021541595459, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 1.1854, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.254283137962128, | |
| "eval_accuracy": 0.7566596229680038, | |
| "eval_loss": 1.4005974531173706, | |
| "eval_runtime": 103.4306, | |
| "eval_samples_per_second": 230.59, | |
| "eval_steps_per_second": 7.213, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.4045686804929365, | |
| "grad_norm": 3.562636613845825, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 1.1894, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.4045686804929365, | |
| "eval_accuracy": 0.7518658362592376, | |
| "eval_loss": 1.4261118173599243, | |
| "eval_runtime": 107.7368, | |
| "eval_samples_per_second": 221.373, | |
| "eval_steps_per_second": 6.924, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.554854223023745, | |
| "grad_norm": 3.0819168090820312, | |
| "learning_rate": 9.15e-05, | |
| "loss": 1.1821, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.554854223023745, | |
| "eval_accuracy": 0.759439375239073, | |
| "eval_loss": 1.3829759359359741, | |
| "eval_runtime": 108.9119, | |
| "eval_samples_per_second": 218.984, | |
| "eval_steps_per_second": 6.85, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.7051397655545535, | |
| "grad_norm": 3.441218852996826, | |
| "learning_rate": 9.1e-05, | |
| "loss": 1.1617, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.7051397655545535, | |
| "eval_accuracy": 0.7595643512285075, | |
| "eval_loss": 1.38172447681427, | |
| "eval_runtime": 107.6971, | |
| "eval_samples_per_second": 221.454, | |
| "eval_steps_per_second": 6.927, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.855425308085362, | |
| "grad_norm": 2.9258110523223877, | |
| "learning_rate": 9.05e-05, | |
| "loss": 1.1647, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.855425308085362, | |
| "eval_accuracy": 0.7622641760596893, | |
| "eval_loss": 1.3564621210098267, | |
| "eval_runtime": 108.8704, | |
| "eval_samples_per_second": 219.068, | |
| "eval_steps_per_second": 6.852, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.005710850616171, | |
| "grad_norm": 3.097913980484009, | |
| "learning_rate": 9e-05, | |
| "loss": 1.1543, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.005710850616171, | |
| "eval_accuracy": 0.7637979723391941, | |
| "eval_loss": 1.3542959690093994, | |
| "eval_runtime": 108.5802, | |
| "eval_samples_per_second": 219.653, | |
| "eval_steps_per_second": 6.87, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.1559963931469794, | |
| "grad_norm": 3.168121099472046, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 1.1379, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.1559963931469794, | |
| "eval_accuracy": 0.76408780640708, | |
| "eval_loss": 1.3504241704940796, | |
| "eval_runtime": 108.2586, | |
| "eval_samples_per_second": 220.306, | |
| "eval_steps_per_second": 6.891, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.306281935677788, | |
| "grad_norm": 3.410958766937256, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 1.1254, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.306281935677788, | |
| "eval_accuracy": 0.7652592815735387, | |
| "eval_loss": 1.346474528312683, | |
| "eval_runtime": 107.3669, | |
| "eval_samples_per_second": 222.136, | |
| "eval_steps_per_second": 6.948, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.4565674782085964, | |
| "grad_norm": 2.775892734527588, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 1.1269, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.4565674782085964, | |
| "eval_accuracy": 0.7657911530106338, | |
| "eval_loss": 1.3591572046279907, | |
| "eval_runtime": 106.844, | |
| "eval_samples_per_second": 223.223, | |
| "eval_steps_per_second": 6.982, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.606853020739405, | |
| "grad_norm": 3.029595375061035, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.1247, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.606853020739405, | |
| "eval_accuracy": 0.767305872889515, | |
| "eval_loss": 1.3377037048339844, | |
| "eval_runtime": 108.9561, | |
| "eval_samples_per_second": 218.896, | |
| "eval_steps_per_second": 6.847, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.7571385632702134, | |
| "grad_norm": 3.1967430114746094, | |
| "learning_rate": 8.75e-05, | |
| "loss": 1.1123, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.7571385632702134, | |
| "eval_accuracy": 0.7687083850392965, | |
| "eval_loss": 1.340959906578064, | |
| "eval_runtime": 107.8508, | |
| "eval_samples_per_second": 221.139, | |
| "eval_steps_per_second": 6.917, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.907424105801022, | |
| "grad_norm": 2.859966993331909, | |
| "learning_rate": 8.7e-05, | |
| "loss": 1.112, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.907424105801022, | |
| "eval_accuracy": 0.7696794726604136, | |
| "eval_loss": 1.3237755298614502, | |
| "eval_runtime": 107.6845, | |
| "eval_samples_per_second": 221.48, | |
| "eval_steps_per_second": 6.928, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.057709648331831, | |
| "grad_norm": 2.8421764373779297, | |
| "learning_rate": 8.65e-05, | |
| "loss": 1.0983, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.057709648331831, | |
| "eval_accuracy": 0.7714031663702743, | |
| "eval_loss": 1.3144177198410034, | |
| "eval_runtime": 108.7428, | |
| "eval_samples_per_second": 219.325, | |
| "eval_steps_per_second": 6.86, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.207995190862639, | |
| "grad_norm": 3.1582884788513184, | |
| "learning_rate": 8.6e-05, | |
| "loss": 1.0918, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.207995190862639, | |
| "eval_accuracy": 0.7718766206792639, | |
| "eval_loss": 1.3221428394317627, | |
| "eval_runtime": 108.1645, | |
| "eval_samples_per_second": 220.498, | |
| "eval_steps_per_second": 6.897, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.358280733393448, | |
| "grad_norm": 2.631855010986328, | |
| "learning_rate": 8.55e-05, | |
| "loss": 1.0884, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.358280733393448, | |
| "eval_accuracy": 0.772324790492508, | |
| "eval_loss": 1.3152235746383667, | |
| "eval_runtime": 107.5113, | |
| "eval_samples_per_second": 221.837, | |
| "eval_steps_per_second": 6.939, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.508566275924256, | |
| "grad_norm": 3.242208480834961, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.0837, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.508566275924256, | |
| "eval_accuracy": 0.7737920955341676, | |
| "eval_loss": 1.2969176769256592, | |
| "eval_runtime": 107.6972, | |
| "eval_samples_per_second": 221.454, | |
| "eval_steps_per_second": 6.927, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.658851818455065, | |
| "grad_norm": 3.0691699981689453, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 1.0796, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.658851818455065, | |
| "eval_accuracy": 0.7738074322270785, | |
| "eval_loss": 1.3085857629776, | |
| "eval_runtime": 106.2258, | |
| "eval_samples_per_second": 224.522, | |
| "eval_steps_per_second": 7.023, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.809137360985873, | |
| "grad_norm": 3.258615732192993, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.0714, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.809137360985873, | |
| "eval_accuracy": 0.775101051869418, | |
| "eval_loss": 1.2939372062683105, | |
| "eval_runtime": 106.8715, | |
| "eval_samples_per_second": 223.165, | |
| "eval_steps_per_second": 6.98, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.959422903516682, | |
| "grad_norm": 3.1920101642608643, | |
| "learning_rate": 8.35e-05, | |
| "loss": 1.071, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.959422903516682, | |
| "eval_accuracy": 0.7753841079350378, | |
| "eval_loss": 1.2854809761047363, | |
| "eval_runtime": 106.6266, | |
| "eval_samples_per_second": 223.678, | |
| "eval_steps_per_second": 6.996, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.10970844604749, | |
| "grad_norm": 3.2814955711364746, | |
| "learning_rate": 8.3e-05, | |
| "loss": 1.0568, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.10970844604749, | |
| "eval_accuracy": 0.7761810007493132, | |
| "eval_loss": 1.2916713953018188, | |
| "eval_runtime": 107.4768, | |
| "eval_samples_per_second": 221.908, | |
| "eval_steps_per_second": 6.941, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.259993988578299, | |
| "grad_norm": 3.2327558994293213, | |
| "learning_rate": 8.25e-05, | |
| "loss": 1.0549, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.259993988578299, | |
| "eval_accuracy": 0.7780586564617779, | |
| "eval_loss": 1.2857751846313477, | |
| "eval_runtime": 107.761, | |
| "eval_samples_per_second": 221.323, | |
| "eval_steps_per_second": 6.923, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.410279531109107, | |
| "grad_norm": 3.0951426029205322, | |
| "learning_rate": 8.2e-05, | |
| "loss": 1.0511, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.410279531109107, | |
| "eval_accuracy": 0.7783227014884583, | |
| "eval_loss": 1.2948368787765503, | |
| "eval_runtime": 108.2092, | |
| "eval_samples_per_second": 220.406, | |
| "eval_steps_per_second": 6.894, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.560565073639916, | |
| "grad_norm": 2.639129638671875, | |
| "learning_rate": 8.15e-05, | |
| "loss": 1.0492, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.560565073639916, | |
| "eval_accuracy": 0.7788534810933934, | |
| "eval_loss": 1.2669661045074463, | |
| "eval_runtime": 108.4058, | |
| "eval_samples_per_second": 220.007, | |
| "eval_steps_per_second": 6.882, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.710850616170724, | |
| "grad_norm": 2.4974186420440674, | |
| "learning_rate": 8.1e-05, | |
| "loss": 1.0476, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.710850616170724, | |
| "eval_accuracy": 0.7792994699668719, | |
| "eval_loss": 1.2653881311416626, | |
| "eval_runtime": 108.006, | |
| "eval_samples_per_second": 220.821, | |
| "eval_steps_per_second": 6.907, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.861136158701533, | |
| "grad_norm": 2.6549036502838135, | |
| "learning_rate": 8.05e-05, | |
| "loss": 1.0537, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 5.861136158701533, | |
| "eval_accuracy": 0.781079231841748, | |
| "eval_loss": 1.2602005004882812, | |
| "eval_runtime": 107.8534, | |
| "eval_samples_per_second": 221.133, | |
| "eval_steps_per_second": 6.917, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.011421701232342, | |
| "grad_norm": 3.047539234161377, | |
| "learning_rate": 8e-05, | |
| "loss": 1.038, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.011421701232342, | |
| "eval_accuracy": 0.7802534672325914, | |
| "eval_loss": 1.2705827951431274, | |
| "eval_runtime": 106.7758, | |
| "eval_samples_per_second": 223.365, | |
| "eval_steps_per_second": 6.987, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.16170724376315, | |
| "grad_norm": 2.7509360313415527, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 1.0268, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.16170724376315, | |
| "eval_accuracy": 0.7803480613075008, | |
| "eval_loss": 1.2649264335632324, | |
| "eval_runtime": 107.1559, | |
| "eval_samples_per_second": 222.573, | |
| "eval_steps_per_second": 6.962, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.311992786293959, | |
| "grad_norm": 2.5355842113494873, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 1.0294, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.311992786293959, | |
| "eval_accuracy": 0.7818066507198538, | |
| "eval_loss": 1.250462532043457, | |
| "eval_runtime": 115.6436, | |
| "eval_samples_per_second": 206.237, | |
| "eval_steps_per_second": 6.451, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.462278328824767, | |
| "grad_norm": 2.9398176670074463, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 1.0209, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.462278328824767, | |
| "eval_accuracy": 0.7832408080027191, | |
| "eval_loss": 1.2534795999526978, | |
| "eval_runtime": 107.8449, | |
| "eval_samples_per_second": 221.151, | |
| "eval_steps_per_second": 6.917, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.612563871355576, | |
| "grad_norm": 2.7950327396392822, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.0268, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.612563871355576, | |
| "eval_accuracy": 0.7836778557957977, | |
| "eval_loss": 1.2419434785842896, | |
| "eval_runtime": 107.3738, | |
| "eval_samples_per_second": 222.121, | |
| "eval_steps_per_second": 6.948, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.762849413886384, | |
| "grad_norm": 3.0478243827819824, | |
| "learning_rate": 7.75e-05, | |
| "loss": 1.0233, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 6.762849413886384, | |
| "eval_accuracy": 0.7839343784481503, | |
| "eval_loss": 1.244408369064331, | |
| "eval_runtime": 107.5196, | |
| "eval_samples_per_second": 221.82, | |
| "eval_steps_per_second": 6.938, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 6.913134956417193, | |
| "grad_norm": 3.049609661102295, | |
| "learning_rate": 7.7e-05, | |
| "loss": 1.016, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 6.913134956417193, | |
| "eval_accuracy": 0.7844505307770736, | |
| "eval_loss": 1.2420412302017212, | |
| "eval_runtime": 108.256, | |
| "eval_samples_per_second": 220.311, | |
| "eval_steps_per_second": 6.891, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.063420498948001, | |
| "grad_norm": 3.2929279804229736, | |
| "learning_rate": 7.65e-05, | |
| "loss": 1.0113, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.063420498948001, | |
| "eval_accuracy": 0.7841336670175184, | |
| "eval_loss": 1.2468925714492798, | |
| "eval_runtime": 108.0777, | |
| "eval_samples_per_second": 220.675, | |
| "eval_steps_per_second": 6.902, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.21370604147881, | |
| "grad_norm": 2.5201022624969482, | |
| "learning_rate": 7.6e-05, | |
| "loss": 1.01, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.21370604147881, | |
| "eval_accuracy": 0.7857022835486672, | |
| "eval_loss": 1.2313475608825684, | |
| "eval_runtime": 108.2646, | |
| "eval_samples_per_second": 220.294, | |
| "eval_steps_per_second": 6.891, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.363991584009618, | |
| "grad_norm": 2.926717519760132, | |
| "learning_rate": 7.55e-05, | |
| "loss": 1.0017, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.363991584009618, | |
| "eval_accuracy": 0.7857564674613335, | |
| "eval_loss": 1.2349497079849243, | |
| "eval_runtime": 107.7822, | |
| "eval_samples_per_second": 221.28, | |
| "eval_steps_per_second": 6.921, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.514277126540427, | |
| "grad_norm": 2.6643712520599365, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.9939, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.514277126540427, | |
| "eval_accuracy": 0.7868022880070933, | |
| "eval_loss": 1.2337555885314941, | |
| "eval_runtime": 106.7322, | |
| "eval_samples_per_second": 223.456, | |
| "eval_steps_per_second": 6.989, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.664562669071236, | |
| "grad_norm": 2.679358720779419, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 0.9921, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.664562669071236, | |
| "eval_accuracy": 0.7861627827490367, | |
| "eval_loss": 1.2326431274414062, | |
| "eval_runtime": 107.5896, | |
| "eval_samples_per_second": 221.676, | |
| "eval_steps_per_second": 6.934, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.814848211602044, | |
| "grad_norm": 2.5837836265563965, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.9973, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.814848211602044, | |
| "eval_accuracy": 0.7871569919088683, | |
| "eval_loss": 1.2331918478012085, | |
| "eval_runtime": 107.9543, | |
| "eval_samples_per_second": 220.927, | |
| "eval_steps_per_second": 6.91, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.965133754132852, | |
| "grad_norm": 2.536043882369995, | |
| "learning_rate": 7.35e-05, | |
| "loss": 0.9953, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 7.965133754132852, | |
| "eval_accuracy": 0.7891570577004067, | |
| "eval_loss": 1.2163290977478027, | |
| "eval_runtime": 113.9576, | |
| "eval_samples_per_second": 209.288, | |
| "eval_steps_per_second": 6.546, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.115419296663662, | |
| "grad_norm": 2.690735340118408, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.9847, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.115419296663662, | |
| "eval_accuracy": 0.7888144688528846, | |
| "eval_loss": 1.209494709968567, | |
| "eval_runtime": 116.0276, | |
| "eval_samples_per_second": 205.555, | |
| "eval_steps_per_second": 6.43, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.26570483919447, | |
| "grad_norm": 2.4185330867767334, | |
| "learning_rate": 7.25e-05, | |
| "loss": 0.9797, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.26570483919447, | |
| "eval_accuracy": 0.7891209106681478, | |
| "eval_loss": 1.218719482421875, | |
| "eval_runtime": 114.5016, | |
| "eval_samples_per_second": 208.294, | |
| "eval_steps_per_second": 6.515, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.415990381725278, | |
| "grad_norm": 2.7199230194091797, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.9755, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 8.415990381725278, | |
| "eval_accuracy": 0.7887393254933777, | |
| "eval_loss": 1.2271952629089355, | |
| "eval_runtime": 111.4173, | |
| "eval_samples_per_second": 214.06, | |
| "eval_steps_per_second": 6.696, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 8.566275924256086, | |
| "grad_norm": 2.6658599376678467, | |
| "learning_rate": 7.15e-05, | |
| "loss": 0.9749, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.566275924256086, | |
| "eval_accuracy": 0.7904619110106559, | |
| "eval_loss": 1.217193603515625, | |
| "eval_runtime": 107.3348, | |
| "eval_samples_per_second": 222.202, | |
| "eval_steps_per_second": 6.95, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.716561466786896, | |
| "grad_norm": 2.9954679012298584, | |
| "learning_rate": 7.1e-05, | |
| "loss": 0.9747, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 8.716561466786896, | |
| "eval_accuracy": 0.7904568069124281, | |
| "eval_loss": 1.2054858207702637, | |
| "eval_runtime": 111.9259, | |
| "eval_samples_per_second": 213.087, | |
| "eval_steps_per_second": 6.665, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 8.866847009317704, | |
| "grad_norm": 2.506471872329712, | |
| "learning_rate": 7.05e-05, | |
| "loss": 0.9715, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 8.866847009317704, | |
| "eval_accuracy": 0.790933514493206, | |
| "eval_loss": 1.203903317451477, | |
| "eval_runtime": 106.5362, | |
| "eval_samples_per_second": 223.868, | |
| "eval_steps_per_second": 7.002, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 9.017132551848512, | |
| "grad_norm": 2.67307710647583, | |
| "learning_rate": 7e-05, | |
| "loss": 0.9716, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.017132551848512, | |
| "eval_accuracy": 0.7914492771721259, | |
| "eval_loss": 1.2048789262771606, | |
| "eval_runtime": 121.4256, | |
| "eval_samples_per_second": 196.417, | |
| "eval_steps_per_second": 6.144, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.16741809437932, | |
| "grad_norm": 2.9317779541015625, | |
| "learning_rate": 6.95e-05, | |
| "loss": 0.962, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 9.16741809437932, | |
| "eval_accuracy": 0.7917621280752731, | |
| "eval_loss": 1.199426293373108, | |
| "eval_runtime": 121.1706, | |
| "eval_samples_per_second": 196.83, | |
| "eval_steps_per_second": 6.157, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 9.31770363691013, | |
| "grad_norm": 2.679703712463379, | |
| "learning_rate": 6.9e-05, | |
| "loss": 0.9642, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 9.31770363691013, | |
| "eval_accuracy": 0.7925397227321643, | |
| "eval_loss": 1.193407654762268, | |
| "eval_runtime": 123.2276, | |
| "eval_samples_per_second": 193.544, | |
| "eval_steps_per_second": 6.054, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 9.467989179440938, | |
| "grad_norm": 2.4788730144500732, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 0.9601, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 9.467989179440938, | |
| "eval_accuracy": 0.792334985338317, | |
| "eval_loss": 1.1934560537338257, | |
| "eval_runtime": 119.6088, | |
| "eval_samples_per_second": 199.4, | |
| "eval_steps_per_second": 6.237, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 9.618274721971746, | |
| "grad_norm": 2.8737170696258545, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.9541, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 9.618274721971746, | |
| "eval_accuracy": 0.7933121889138387, | |
| "eval_loss": 1.1921508312225342, | |
| "eval_runtime": 124.8462, | |
| "eval_samples_per_second": 191.035, | |
| "eval_steps_per_second": 5.975, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 9.768560264502554, | |
| "grad_norm": 2.7593533992767334, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 0.9599, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 9.768560264502554, | |
| "eval_accuracy": 0.7930929148774838, | |
| "eval_loss": 1.2042547464370728, | |
| "eval_runtime": 124.9577, | |
| "eval_samples_per_second": 190.865, | |
| "eval_steps_per_second": 5.97, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 9.918845807033364, | |
| "grad_norm": 2.389718532562256, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.9649, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 9.918845807033364, | |
| "eval_accuracy": 0.7939532145922783, | |
| "eval_loss": 1.1914194822311401, | |
| "eval_runtime": 120.967, | |
| "eval_samples_per_second": 197.161, | |
| "eval_steps_per_second": 6.167, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 10.069131349564172, | |
| "grad_norm": 2.846874475479126, | |
| "learning_rate": 6.65e-05, | |
| "loss": 0.9467, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 10.069131349564172, | |
| "eval_accuracy": 0.7947739980138828, | |
| "eval_loss": 1.1852329969406128, | |
| "eval_runtime": 120.8513, | |
| "eval_samples_per_second": 197.35, | |
| "eval_steps_per_second": 6.173, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 10.21941689209498, | |
| "grad_norm": 2.58475399017334, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.9406, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 10.21941689209498, | |
| "eval_accuracy": 0.7943515220267853, | |
| "eval_loss": 1.1843186616897583, | |
| "eval_runtime": 121.8305, | |
| "eval_samples_per_second": 195.764, | |
| "eval_steps_per_second": 6.123, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 10.36970243462579, | |
| "grad_norm": 3.003615140914917, | |
| "learning_rate": 6.55e-05, | |
| "loss": 0.9395, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 10.36970243462579, | |
| "eval_accuracy": 0.7948140182397172, | |
| "eval_loss": 1.1831063032150269, | |
| "eval_runtime": 120.2856, | |
| "eval_samples_per_second": 198.278, | |
| "eval_steps_per_second": 6.202, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 10.519987977156598, | |
| "grad_norm": 2.6997032165527344, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.9338, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 10.519987977156598, | |
| "eval_accuracy": 0.7955270740039722, | |
| "eval_loss": 1.1888868808746338, | |
| "eval_runtime": 121.3291, | |
| "eval_samples_per_second": 196.573, | |
| "eval_steps_per_second": 6.149, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 10.670273519687406, | |
| "grad_norm": 2.5620720386505127, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 0.9413, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 10.670273519687406, | |
| "eval_accuracy": 0.7960009228127132, | |
| "eval_loss": 1.1771271228790283, | |
| "eval_runtime": 116.3789, | |
| "eval_samples_per_second": 204.934, | |
| "eval_steps_per_second": 6.41, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 10.820559062218214, | |
| "grad_norm": 2.4902310371398926, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.941, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 10.820559062218214, | |
| "eval_accuracy": 0.7952424860773146, | |
| "eval_loss": 1.176774263381958, | |
| "eval_runtime": 116.6512, | |
| "eval_samples_per_second": 204.456, | |
| "eval_steps_per_second": 6.395, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 10.970844604749024, | |
| "grad_norm": 2.6512043476104736, | |
| "learning_rate": 6.35e-05, | |
| "loss": 0.9371, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 10.970844604749024, | |
| "eval_accuracy": 0.796440942176153, | |
| "eval_loss": 1.1723679304122925, | |
| "eval_runtime": 119.8979, | |
| "eval_samples_per_second": 198.919, | |
| "eval_steps_per_second": 6.222, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 11.121130147279832, | |
| "grad_norm": 2.6038336753845215, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.9293, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 11.121130147279832, | |
| "eval_accuracy": 0.7967239525153044, | |
| "eval_loss": 1.1750506162643433, | |
| "eval_runtime": 120.4764, | |
| "eval_samples_per_second": 197.964, | |
| "eval_steps_per_second": 6.192, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 11.27141568981064, | |
| "grad_norm": 2.5120317935943604, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.9279, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 11.27141568981064, | |
| "eval_accuracy": 0.7974463986990397, | |
| "eval_loss": 1.1721601486206055, | |
| "eval_runtime": 115.5123, | |
| "eval_samples_per_second": 206.471, | |
| "eval_steps_per_second": 6.458, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 11.421701232341448, | |
| "grad_norm": 2.6776065826416016, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.9273, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 11.421701232341448, | |
| "eval_accuracy": 0.7977723054491466, | |
| "eval_loss": 1.1790024042129517, | |
| "eval_runtime": 118.0413, | |
| "eval_samples_per_second": 202.048, | |
| "eval_steps_per_second": 6.32, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 11.571986774872258, | |
| "grad_norm": 2.473292589187622, | |
| "learning_rate": 6.15e-05, | |
| "loss": 0.9176, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 11.571986774872258, | |
| "eval_accuracy": 0.7978240654330204, | |
| "eval_loss": 1.1769903898239136, | |
| "eval_runtime": 119.5414, | |
| "eval_samples_per_second": 199.512, | |
| "eval_steps_per_second": 6.241, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 11.722272317403066, | |
| "grad_norm": 2.573493242263794, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.918, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 11.722272317403066, | |
| "eval_accuracy": 0.7978464067774307, | |
| "eval_loss": 1.1745543479919434, | |
| "eval_runtime": 118.8031, | |
| "eval_samples_per_second": 200.752, | |
| "eval_steps_per_second": 6.279, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 11.872557859933874, | |
| "grad_norm": 2.496293067932129, | |
| "learning_rate": 6.05e-05, | |
| "loss": 0.9209, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 11.872557859933874, | |
| "eval_accuracy": 0.7988222222591843, | |
| "eval_loss": 1.1572139263153076, | |
| "eval_runtime": 119.3596, | |
| "eval_samples_per_second": 199.816, | |
| "eval_steps_per_second": 6.25, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 12.022843402464684, | |
| "grad_norm": 2.8805453777313232, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9204, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 12.022843402464684, | |
| "eval_accuracy": 0.799307208811464, | |
| "eval_loss": 1.1740858554840088, | |
| "eval_runtime": 118.6109, | |
| "eval_samples_per_second": 201.078, | |
| "eval_steps_per_second": 6.289, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 12.173128944995492, | |
| "grad_norm": 2.6088273525238037, | |
| "learning_rate": 5.95e-05, | |
| "loss": 0.9149, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 12.173128944995492, | |
| "eval_accuracy": 0.7991013549531508, | |
| "eval_loss": 1.1727755069732666, | |
| "eval_runtime": 117.7954, | |
| "eval_samples_per_second": 202.47, | |
| "eval_steps_per_second": 6.333, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 12.3234144875263, | |
| "grad_norm": 2.426567316055298, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.9009, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 12.3234144875263, | |
| "eval_accuracy": 0.798481914968918, | |
| "eval_loss": 1.1672251224517822, | |
| "eval_runtime": 119.0757, | |
| "eval_samples_per_second": 200.293, | |
| "eval_steps_per_second": 6.265, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 12.473700030057108, | |
| "grad_norm": 2.687640428543091, | |
| "learning_rate": 5.85e-05, | |
| "loss": 0.9094, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 12.473700030057108, | |
| "eval_accuracy": 0.7998230800562746, | |
| "eval_loss": 1.1598495244979858, | |
| "eval_runtime": 118.9102, | |
| "eval_samples_per_second": 200.572, | |
| "eval_steps_per_second": 6.274, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 12.623985572587918, | |
| "grad_norm": 3.0333776473999023, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.9149, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 12.623985572587918, | |
| "eval_accuracy": 0.7999625316919439, | |
| "eval_loss": 1.171325445175171, | |
| "eval_runtime": 115.3742, | |
| "eval_samples_per_second": 206.719, | |
| "eval_steps_per_second": 6.466, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 12.774271115118726, | |
| "grad_norm": 2.299436569213867, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.9068, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 12.774271115118726, | |
| "eval_accuracy": 0.8010648350504735, | |
| "eval_loss": 1.1537537574768066, | |
| "eval_runtime": 118.7443, | |
| "eval_samples_per_second": 200.852, | |
| "eval_steps_per_second": 6.282, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 12.924556657649534, | |
| "grad_norm": 2.7340447902679443, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.9099, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 12.924556657649534, | |
| "eval_accuracy": 0.8012578176921267, | |
| "eval_loss": 1.1426852941513062, | |
| "eval_runtime": 120.2664, | |
| "eval_samples_per_second": 198.31, | |
| "eval_steps_per_second": 6.203, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 13.074842200180342, | |
| "grad_norm": 2.6585094928741455, | |
| "learning_rate": 5.65e-05, | |
| "loss": 0.8951, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 13.074842200180342, | |
| "eval_accuracy": 0.8005615361847466, | |
| "eval_loss": 1.1578710079193115, | |
| "eval_runtime": 115.0048, | |
| "eval_samples_per_second": 207.383, | |
| "eval_steps_per_second": 6.487, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 13.225127742711152, | |
| "grad_norm": 2.6981582641601562, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.8886, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 13.225127742711152, | |
| "eval_accuracy": 0.8010340860790154, | |
| "eval_loss": 1.1602416038513184, | |
| "eval_runtime": 113.2435, | |
| "eval_samples_per_second": 210.608, | |
| "eval_steps_per_second": 6.588, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 13.37541328524196, | |
| "grad_norm": 2.6016407012939453, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 0.9057, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 13.37541328524196, | |
| "eval_accuracy": 0.8021396084503188, | |
| "eval_loss": 1.1406522989273071, | |
| "eval_runtime": 115.7578, | |
| "eval_samples_per_second": 206.034, | |
| "eval_steps_per_second": 6.444, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 13.525698827772768, | |
| "grad_norm": 2.693239688873291, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.8921, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 13.525698827772768, | |
| "eval_accuracy": 0.801952397809383, | |
| "eval_loss": 1.1361913681030273, | |
| "eval_runtime": 112.7804, | |
| "eval_samples_per_second": 211.473, | |
| "eval_steps_per_second": 6.615, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 13.675984370303578, | |
| "grad_norm": 2.386279582977295, | |
| "learning_rate": 5.45e-05, | |
| "loss": 0.8942, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 13.675984370303578, | |
| "eval_accuracy": 0.8024716769653207, | |
| "eval_loss": 1.1441001892089844, | |
| "eval_runtime": 114.7396, | |
| "eval_samples_per_second": 207.862, | |
| "eval_steps_per_second": 6.502, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 13.826269912834386, | |
| "grad_norm": 2.3237531185150146, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.8939, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 13.826269912834386, | |
| "eval_accuracy": 0.8025017531571386, | |
| "eval_loss": 1.141733169555664, | |
| "eval_runtime": 112.3505, | |
| "eval_samples_per_second": 212.282, | |
| "eval_steps_per_second": 6.64, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 13.976555455365194, | |
| "grad_norm": 2.463498592376709, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 0.891, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 13.976555455365194, | |
| "eval_accuracy": 0.803545043091428, | |
| "eval_loss": 1.144685983657837, | |
| "eval_runtime": 114.632, | |
| "eval_samples_per_second": 208.057, | |
| "eval_steps_per_second": 6.508, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 14.126840997896002, | |
| "grad_norm": 2.7393481731414795, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 0.8846, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 14.126840997896002, | |
| "eval_accuracy": 0.8038034340555349, | |
| "eval_loss": 1.1320561170578003, | |
| "eval_runtime": 112.5915, | |
| "eval_samples_per_second": 211.828, | |
| "eval_steps_per_second": 6.626, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 14.277126540426812, | |
| "grad_norm": 2.776505708694458, | |
| "learning_rate": 5.25e-05, | |
| "loss": 0.8841, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 14.277126540426812, | |
| "eval_accuracy": 0.8034705833910766, | |
| "eval_loss": 1.1436700820922852, | |
| "eval_runtime": 114.9721, | |
| "eval_samples_per_second": 207.442, | |
| "eval_steps_per_second": 6.489, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 14.42741208295762, | |
| "grad_norm": 2.536817789077759, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.8811, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 14.42741208295762, | |
| "eval_accuracy": 0.8044537392561116, | |
| "eval_loss": 1.133318305015564, | |
| "eval_runtime": 113.5045, | |
| "eval_samples_per_second": 210.124, | |
| "eval_steps_per_second": 6.572, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 14.577697625488428, | |
| "grad_norm": 2.524203062057495, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 0.8778, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 14.577697625488428, | |
| "eval_accuracy": 0.8039248049976617, | |
| "eval_loss": 1.1258032321929932, | |
| "eval_runtime": 114.9378, | |
| "eval_samples_per_second": 207.504, | |
| "eval_steps_per_second": 6.49, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 14.727983168019236, | |
| "grad_norm": 2.6129727363586426, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 0.8765, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 14.727983168019236, | |
| "eval_accuracy": 0.8044657041630938, | |
| "eval_loss": 1.1352417469024658, | |
| "eval_runtime": 114.1357, | |
| "eval_samples_per_second": 208.962, | |
| "eval_steps_per_second": 6.536, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 14.878268710550046, | |
| "grad_norm": 2.6218881607055664, | |
| "learning_rate": 5.05e-05, | |
| "loss": 0.876, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 14.878268710550046, | |
| "eval_accuracy": 0.8049223412984352, | |
| "eval_loss": 1.131935954093933, | |
| "eval_runtime": 114.4278, | |
| "eval_samples_per_second": 208.428, | |
| "eval_steps_per_second": 6.519, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 15.028554253080854, | |
| "grad_norm": 2.654724597930908, | |
| "learning_rate": 5e-05, | |
| "loss": 0.879, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 15.028554253080854, | |
| "eval_accuracy": 0.8056327032697777, | |
| "eval_loss": 1.1161094903945923, | |
| "eval_runtime": 112.2273, | |
| "eval_samples_per_second": 212.515, | |
| "eval_steps_per_second": 6.647, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 15.178839795611662, | |
| "grad_norm": 2.569153308868408, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.8674, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 15.178839795611662, | |
| "eval_accuracy": 0.8048139780991767, | |
| "eval_loss": 1.1278635263442993, | |
| "eval_runtime": 114.7277, | |
| "eval_samples_per_second": 207.884, | |
| "eval_steps_per_second": 6.502, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 15.32912533814247, | |
| "grad_norm": 2.5382237434387207, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.8676, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 15.32912533814247, | |
| "eval_accuracy": 0.8064227363767525, | |
| "eval_loss": 1.132450819015503, | |
| "eval_runtime": 107.2478, | |
| "eval_samples_per_second": 222.382, | |
| "eval_steps_per_second": 6.956, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 15.47941088067328, | |
| "grad_norm": 2.480746030807495, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.867, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 15.47941088067328, | |
| "eval_accuracy": 0.806509341887431, | |
| "eval_loss": 1.1208624839782715, | |
| "eval_runtime": 107.5739, | |
| "eval_samples_per_second": 221.708, | |
| "eval_steps_per_second": 6.935, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 15.629696423204088, | |
| "grad_norm": 2.4417829513549805, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.8648, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 15.629696423204088, | |
| "eval_accuracy": 0.80664691560334, | |
| "eval_loss": 1.123421549797058, | |
| "eval_runtime": 107.0763, | |
| "eval_samples_per_second": 222.738, | |
| "eval_steps_per_second": 6.967, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 15.779981965734896, | |
| "grad_norm": 2.3048479557037354, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.87, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 15.779981965734896, | |
| "eval_accuracy": 0.8067313801594327, | |
| "eval_loss": 1.1155991554260254, | |
| "eval_runtime": 108.1545, | |
| "eval_samples_per_second": 220.518, | |
| "eval_steps_per_second": 6.898, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 15.930267508265704, | |
| "grad_norm": 2.6421682834625244, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.8694, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 15.930267508265704, | |
| "eval_accuracy": 0.8066291950596075, | |
| "eval_loss": 1.1215757131576538, | |
| "eval_runtime": 109.1629, | |
| "eval_samples_per_second": 218.481, | |
| "eval_steps_per_second": 6.834, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 16.080553050796514, | |
| "grad_norm": 2.4313910007476807, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.8668, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 16.080553050796514, | |
| "eval_accuracy": 0.8079721270128156, | |
| "eval_loss": 1.1246066093444824, | |
| "eval_runtime": 108.472, | |
| "eval_samples_per_second": 219.872, | |
| "eval_steps_per_second": 6.877, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 16.230838593327324, | |
| "grad_norm": 2.2622108459472656, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.8603, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 16.230838593327324, | |
| "eval_accuracy": 0.8070964068446762, | |
| "eval_loss": 1.1118344068527222, | |
| "eval_runtime": 107.6522, | |
| "eval_samples_per_second": 221.547, | |
| "eval_steps_per_second": 6.93, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 16.38112413585813, | |
| "grad_norm": 2.355970621109009, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.8557, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 16.38112413585813, | |
| "eval_accuracy": 0.8079940925101587, | |
| "eval_loss": 1.1261143684387207, | |
| "eval_runtime": 107.1531, | |
| "eval_samples_per_second": 222.579, | |
| "eval_steps_per_second": 6.962, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 16.53140967838894, | |
| "grad_norm": 2.4477593898773193, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8521, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 16.53140967838894, | |
| "eval_accuracy": 0.8084604086659483, | |
| "eval_loss": 1.1086950302124023, | |
| "eval_runtime": 106.6462, | |
| "eval_samples_per_second": 223.637, | |
| "eval_steps_per_second": 6.995, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 16.681695220919746, | |
| "grad_norm": 3.1719090938568115, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.8583, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 16.681695220919746, | |
| "eval_accuracy": 0.8085530062732024, | |
| "eval_loss": 1.1111265420913696, | |
| "eval_runtime": 107.453, | |
| "eval_samples_per_second": 221.958, | |
| "eval_steps_per_second": 6.943, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 16.831980763450556, | |
| "grad_norm": 2.867218494415283, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.8509, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 16.831980763450556, | |
| "eval_accuracy": 0.8083466491211581, | |
| "eval_loss": 1.1217560768127441, | |
| "eval_runtime": 107.5379, | |
| "eval_samples_per_second": 221.782, | |
| "eval_steps_per_second": 6.937, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 16.982266305981366, | |
| "grad_norm": 2.421694040298462, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.8591, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 16.982266305981366, | |
| "eval_accuracy": 0.808431427546503, | |
| "eval_loss": 1.1109532117843628, | |
| "eval_runtime": 108.5465, | |
| "eval_samples_per_second": 219.721, | |
| "eval_steps_per_second": 6.873, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 17.132551848512172, | |
| "grad_norm": 2.6324167251586914, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.8417, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 17.132551848512172, | |
| "eval_accuracy": 0.8085933337773042, | |
| "eval_loss": 1.1125506162643433, | |
| "eval_runtime": 108.0877, | |
| "eval_samples_per_second": 220.654, | |
| "eval_steps_per_second": 6.902, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 17.282837391042982, | |
| "grad_norm": 2.448883056640625, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.8511, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 17.282837391042982, | |
| "eval_accuracy": 0.8093454793356915, | |
| "eval_loss": 1.1108651161193848, | |
| "eval_runtime": 108.0036, | |
| "eval_samples_per_second": 220.826, | |
| "eval_steps_per_second": 6.907, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 17.43312293357379, | |
| "grad_norm": 2.647814989089966, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.8472, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 17.43312293357379, | |
| "eval_accuracy": 0.8097202877059791, | |
| "eval_loss": 1.1124111413955688, | |
| "eval_runtime": 107.9075, | |
| "eval_samples_per_second": 221.023, | |
| "eval_steps_per_second": 6.913, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 17.583408476104598, | |
| "grad_norm": 2.3744354248046875, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.8381, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 17.583408476104598, | |
| "eval_accuracy": 0.8095948479389424, | |
| "eval_loss": 1.1056084632873535, | |
| "eval_runtime": 107.6903, | |
| "eval_samples_per_second": 221.468, | |
| "eval_steps_per_second": 6.927, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 17.733694018635408, | |
| "grad_norm": 2.3030834197998047, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.8474, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 17.733694018635408, | |
| "eval_accuracy": 0.8102527131048097, | |
| "eval_loss": 1.1011704206466675, | |
| "eval_runtime": 107.5969, | |
| "eval_samples_per_second": 221.661, | |
| "eval_steps_per_second": 6.933, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 17.883979561166214, | |
| "grad_norm": 2.610208749771118, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.8456, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 17.883979561166214, | |
| "eval_accuracy": 0.8108135572608609, | |
| "eval_loss": 1.097122073173523, | |
| "eval_runtime": 106.6046, | |
| "eval_samples_per_second": 223.724, | |
| "eval_steps_per_second": 6.998, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 18.034265103697024, | |
| "grad_norm": 2.491633176803589, | |
| "learning_rate": 4e-05, | |
| "loss": 0.8367, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 18.034265103697024, | |
| "eval_accuracy": 0.8112650064622723, | |
| "eval_loss": 1.0913532972335815, | |
| "eval_runtime": 107.2029, | |
| "eval_samples_per_second": 222.475, | |
| "eval_steps_per_second": 6.959, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 18.184550646227834, | |
| "grad_norm": 2.270582437515259, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.8336, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 18.184550646227834, | |
| "eval_accuracy": 0.8104727212172935, | |
| "eval_loss": 1.110021710395813, | |
| "eval_runtime": 106.0993, | |
| "eval_samples_per_second": 224.789, | |
| "eval_steps_per_second": 7.031, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 18.33483618875864, | |
| "grad_norm": 2.4716830253601074, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.8404, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 18.33483618875864, | |
| "eval_accuracy": 0.8114022782578146, | |
| "eval_loss": 1.0959724187850952, | |
| "eval_runtime": 107.1337, | |
| "eval_samples_per_second": 222.619, | |
| "eval_steps_per_second": 6.963, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 18.48512173128945, | |
| "grad_norm": 2.679825782775879, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.8365, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 18.48512173128945, | |
| "eval_accuracy": 0.8112698414020383, | |
| "eval_loss": 1.0895620584487915, | |
| "eval_runtime": 107.9425, | |
| "eval_samples_per_second": 220.951, | |
| "eval_steps_per_second": 6.911, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 18.63540727382026, | |
| "grad_norm": 2.5028414726257324, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.8308, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 18.63540727382026, | |
| "eval_accuracy": 0.8112445219475855, | |
| "eval_loss": 1.098541498184204, | |
| "eval_runtime": 108.1696, | |
| "eval_samples_per_second": 220.487, | |
| "eval_steps_per_second": 6.897, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 18.785692816351066, | |
| "grad_norm": 2.4303503036499023, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.8308, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 18.785692816351066, | |
| "eval_accuracy": 0.8120511999201301, | |
| "eval_loss": 1.0968284606933594, | |
| "eval_runtime": 107.5288, | |
| "eval_samples_per_second": 221.801, | |
| "eval_steps_per_second": 6.938, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 18.935978358881876, | |
| "grad_norm": 2.9847395420074463, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.8312, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 18.935978358881876, | |
| "eval_accuracy": 0.8121808312724796, | |
| "eval_loss": 1.0884159803390503, | |
| "eval_runtime": 107.9223, | |
| "eval_samples_per_second": 220.992, | |
| "eval_steps_per_second": 6.912, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 19.086263901412686, | |
| "grad_norm": 2.302724838256836, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.8279, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 19.086263901412686, | |
| "eval_accuracy": 0.8127240285256392, | |
| "eval_loss": 1.0900928974151611, | |
| "eval_runtime": 106.9119, | |
| "eval_samples_per_second": 223.081, | |
| "eval_steps_per_second": 6.978, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 19.236549443943492, | |
| "grad_norm": 2.4187684059143066, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.8219, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 19.236549443943492, | |
| "eval_accuracy": 0.8129403972187366, | |
| "eval_loss": 1.0872172117233276, | |
| "eval_runtime": 109.8336, | |
| "eval_samples_per_second": 217.147, | |
| "eval_steps_per_second": 6.792, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 19.3868349864743, | |
| "grad_norm": 2.740501880645752, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.8266, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 19.3868349864743, | |
| "eval_accuracy": 0.8114303723800741, | |
| "eval_loss": 1.0987831354141235, | |
| "eval_runtime": 109.7168, | |
| "eval_samples_per_second": 217.378, | |
| "eval_steps_per_second": 6.799, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 19.537120529005108, | |
| "grad_norm": 2.326366424560547, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.8255, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 19.537120529005108, | |
| "eval_accuracy": 0.8124286882301512, | |
| "eval_loss": 1.0885217189788818, | |
| "eval_runtime": 114.778, | |
| "eval_samples_per_second": 207.792, | |
| "eval_steps_per_second": 6.5, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 19.687406071535918, | |
| "grad_norm": 3.1694321632385254, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.8185, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 19.687406071535918, | |
| "eval_accuracy": 0.8129749253547676, | |
| "eval_loss": 1.0836535692214966, | |
| "eval_runtime": 114.5951, | |
| "eval_samples_per_second": 208.124, | |
| "eval_steps_per_second": 6.51, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 19.837691614066728, | |
| "grad_norm": 2.6415817737579346, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.8219, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 19.837691614066728, | |
| "eval_accuracy": 0.8122496169754481, | |
| "eval_loss": 1.0868114233016968, | |
| "eval_runtime": 113.7304, | |
| "eval_samples_per_second": 209.707, | |
| "eval_steps_per_second": 6.559, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 19.987977156597534, | |
| "grad_norm": 2.567044496536255, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.8214, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 19.987977156597534, | |
| "eval_accuracy": 0.8135328500351103, | |
| "eval_loss": 1.0849970579147339, | |
| "eval_runtime": 114.2096, | |
| "eval_samples_per_second": 208.827, | |
| "eval_steps_per_second": 6.532, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 20.138262699128344, | |
| "grad_norm": 2.475660562515259, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.8123, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 20.138262699128344, | |
| "eval_accuracy": 0.8127171094527156, | |
| "eval_loss": 1.0919703245162964, | |
| "eval_runtime": 110.4581, | |
| "eval_samples_per_second": 215.919, | |
| "eval_steps_per_second": 6.754, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 20.288548241659154, | |
| "grad_norm": 2.9205057621002197, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.8146, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 20.288548241659154, | |
| "eval_accuracy": 0.8139352319239987, | |
| "eval_loss": 1.0828359127044678, | |
| "eval_runtime": 110.5729, | |
| "eval_samples_per_second": 215.695, | |
| "eval_steps_per_second": 6.747, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 20.43883378418996, | |
| "grad_norm": 2.775470018386841, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.8117, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 20.43883378418996, | |
| "eval_accuracy": 0.8144743916913453, | |
| "eval_loss": 1.0843496322631836, | |
| "eval_runtime": 110.9504, | |
| "eval_samples_per_second": 214.961, | |
| "eval_steps_per_second": 6.724, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 20.58911932672077, | |
| "grad_norm": 2.8596460819244385, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.8142, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 20.58911932672077, | |
| "eval_accuracy": 0.8145867898001746, | |
| "eval_loss": 1.0775424242019653, | |
| "eval_runtime": 110.2042, | |
| "eval_samples_per_second": 216.416, | |
| "eval_steps_per_second": 6.769, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 20.73940486925158, | |
| "grad_norm": 2.5671284198760986, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.8176, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 20.73940486925158, | |
| "eval_accuracy": 0.8140961234786387, | |
| "eval_loss": 1.0756142139434814, | |
| "eval_runtime": 110.7907, | |
| "eval_samples_per_second": 215.271, | |
| "eval_steps_per_second": 6.733, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 20.889690411782386, | |
| "grad_norm": 2.549713373184204, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.813, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 20.889690411782386, | |
| "eval_accuracy": 0.8145543791131244, | |
| "eval_loss": 1.0741270780563354, | |
| "eval_runtime": 113.2044, | |
| "eval_samples_per_second": 210.681, | |
| "eval_steps_per_second": 6.59, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 21.039975954313196, | |
| "grad_norm": 2.433366060256958, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8044, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 21.039975954313196, | |
| "eval_accuracy": 0.8138238331972251, | |
| "eval_loss": 1.0890129804611206, | |
| "eval_runtime": 113.5064, | |
| "eval_samples_per_second": 210.12, | |
| "eval_steps_per_second": 6.572, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 21.190261496844002, | |
| "grad_norm": 2.672717571258545, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.8034, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 21.190261496844002, | |
| "eval_accuracy": 0.8146000723609511, | |
| "eval_loss": 1.0757637023925781, | |
| "eval_runtime": 109.4526, | |
| "eval_samples_per_second": 217.902, | |
| "eval_steps_per_second": 6.816, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 21.340547039374812, | |
| "grad_norm": 2.7107882499694824, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.8007, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 21.340547039374812, | |
| "eval_accuracy": 0.815981095746543, | |
| "eval_loss": 1.074793815612793, | |
| "eval_runtime": 113.0561, | |
| "eval_samples_per_second": 210.957, | |
| "eval_steps_per_second": 6.598, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 21.49083258190562, | |
| "grad_norm": 2.419224262237549, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.8009, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 21.49083258190562, | |
| "eval_accuracy": 0.8157551774698867, | |
| "eval_loss": 1.0713990926742554, | |
| "eval_runtime": 114.1603, | |
| "eval_samples_per_second": 208.917, | |
| "eval_steps_per_second": 6.535, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 21.641118124436428, | |
| "grad_norm": 2.68849778175354, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.8042, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 21.641118124436428, | |
| "eval_accuracy": 0.816240857712766, | |
| "eval_loss": 1.0804829597473145, | |
| "eval_runtime": 113.3941, | |
| "eval_samples_per_second": 210.328, | |
| "eval_steps_per_second": 6.579, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 21.791403666967238, | |
| "grad_norm": 2.3715298175811768, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.7971, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 21.791403666967238, | |
| "eval_accuracy": 0.8154816358162141, | |
| "eval_loss": 1.0725679397583008, | |
| "eval_runtime": 114.0456, | |
| "eval_samples_per_second": 209.127, | |
| "eval_steps_per_second": 6.541, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 21.941689209498048, | |
| "grad_norm": 2.6499125957489014, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.807, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 21.941689209498048, | |
| "eval_accuracy": 0.8168061183380335, | |
| "eval_loss": 1.0588874816894531, | |
| "eval_runtime": 112.5964, | |
| "eval_samples_per_second": 211.819, | |
| "eval_steps_per_second": 6.625, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 22.091974752028854, | |
| "grad_norm": 2.766451358795166, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.7973, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 22.091974752028854, | |
| "eval_accuracy": 0.8168862140868284, | |
| "eval_loss": 1.063796877861023, | |
| "eval_runtime": 113.9911, | |
| "eval_samples_per_second": 209.227, | |
| "eval_steps_per_second": 6.544, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 22.242260294559664, | |
| "grad_norm": 2.5453474521636963, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.7915, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 22.242260294559664, | |
| "eval_accuracy": 0.8162707303528417, | |
| "eval_loss": 1.068402886390686, | |
| "eval_runtime": 113.1917, | |
| "eval_samples_per_second": 210.704, | |
| "eval_steps_per_second": 6.591, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 22.392545837090474, | |
| "grad_norm": 2.449525833129883, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.8019, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 22.392545837090474, | |
| "eval_accuracy": 0.8168465033060288, | |
| "eval_loss": 1.071266531944275, | |
| "eval_runtime": 114.1644, | |
| "eval_samples_per_second": 208.909, | |
| "eval_steps_per_second": 6.534, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 22.54283137962128, | |
| "grad_norm": 2.782717704772949, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7959, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 22.54283137962128, | |
| "eval_accuracy": 0.8176195224595183, | |
| "eval_loss": 1.0642682313919067, | |
| "eval_runtime": 112.938, | |
| "eval_samples_per_second": 211.178, | |
| "eval_steps_per_second": 6.605, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 22.69311692215209, | |
| "grad_norm": 2.754309892654419, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.7905, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 22.69311692215209, | |
| "eval_accuracy": 0.8177573368082611, | |
| "eval_loss": 1.0715863704681396, | |
| "eval_runtime": 113.2687, | |
| "eval_samples_per_second": 210.561, | |
| "eval_steps_per_second": 6.586, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 22.843402464682896, | |
| "grad_norm": 2.665132999420166, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.7894, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 22.843402464682896, | |
| "eval_accuracy": 0.8184662117931626, | |
| "eval_loss": 1.0566191673278809, | |
| "eval_runtime": 114.3215, | |
| "eval_samples_per_second": 208.622, | |
| "eval_steps_per_second": 6.525, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 22.993688007213706, | |
| "grad_norm": 2.2912895679473877, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.789, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 22.993688007213706, | |
| "eval_accuracy": 0.8173126447012466, | |
| "eval_loss": 1.0590641498565674, | |
| "eval_runtime": 114.0949, | |
| "eval_samples_per_second": 209.037, | |
| "eval_steps_per_second": 6.538, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 23.143973549744516, | |
| "grad_norm": 2.7320480346679688, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.7859, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 23.143973549744516, | |
| "eval_accuracy": 0.8181692377590023, | |
| "eval_loss": 1.0568209886550903, | |
| "eval_runtime": 114.0712, | |
| "eval_samples_per_second": 209.08, | |
| "eval_steps_per_second": 6.54, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 23.294259092275322, | |
| "grad_norm": 2.5960936546325684, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.7894, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 23.294259092275322, | |
| "eval_accuracy": 0.8178769275591534, | |
| "eval_loss": 1.061540961265564, | |
| "eval_runtime": 113.9113, | |
| "eval_samples_per_second": 209.373, | |
| "eval_steps_per_second": 6.549, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 23.44454463480613, | |
| "grad_norm": 2.435558557510376, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.7887, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 23.44454463480613, | |
| "eval_accuracy": 0.8177420807085761, | |
| "eval_loss": 1.0617201328277588, | |
| "eval_runtime": 113.4776, | |
| "eval_samples_per_second": 210.174, | |
| "eval_steps_per_second": 6.574, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 23.59483017733694, | |
| "grad_norm": 2.3692948818206787, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.7826, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 23.59483017733694, | |
| "eval_accuracy": 0.8180053577064287, | |
| "eval_loss": 1.0567620992660522, | |
| "eval_runtime": 113.7057, | |
| "eval_samples_per_second": 209.752, | |
| "eval_steps_per_second": 6.561, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 23.745115719867748, | |
| "grad_norm": 2.5609283447265625, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.7885, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 23.745115719867748, | |
| "eval_accuracy": 0.8188002648627605, | |
| "eval_loss": 1.0632256269454956, | |
| "eval_runtime": 113.6739, | |
| "eval_samples_per_second": 209.811, | |
| "eval_steps_per_second": 6.563, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 23.895401262398558, | |
| "grad_norm": 2.3372645378112793, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.7883, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 23.895401262398558, | |
| "eval_accuracy": 0.8187176321623402, | |
| "eval_loss": 1.066706657409668, | |
| "eval_runtime": 114.0886, | |
| "eval_samples_per_second": 209.048, | |
| "eval_steps_per_second": 6.539, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 24.045686804929367, | |
| "grad_norm": 2.3112826347351074, | |
| "learning_rate": 2e-05, | |
| "loss": 0.783, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 24.045686804929367, | |
| "eval_accuracy": 0.818846919349616, | |
| "eval_loss": 1.0611591339111328, | |
| "eval_runtime": 113.1339, | |
| "eval_samples_per_second": 210.812, | |
| "eval_steps_per_second": 6.594, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 24.195972347460174, | |
| "grad_norm": 2.514390230178833, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.7804, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 24.195972347460174, | |
| "eval_accuracy": 0.818812757678464, | |
| "eval_loss": 1.0573077201843262, | |
| "eval_runtime": 114.2005, | |
| "eval_samples_per_second": 208.843, | |
| "eval_steps_per_second": 6.532, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 24.346257889990984, | |
| "grad_norm": 2.4737348556518555, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.7811, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 24.346257889990984, | |
| "eval_accuracy": 0.8192913294066332, | |
| "eval_loss": 1.0586949586868286, | |
| "eval_runtime": 113.7459, | |
| "eval_samples_per_second": 209.678, | |
| "eval_steps_per_second": 6.558, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 24.49654343252179, | |
| "grad_norm": 2.288757562637329, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.7767, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 24.49654343252179, | |
| "eval_accuracy": 0.8197264062916556, | |
| "eval_loss": 1.0525128841400146, | |
| "eval_runtime": 113.6132, | |
| "eval_samples_per_second": 209.923, | |
| "eval_steps_per_second": 6.566, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 24.6468289750526, | |
| "grad_norm": 2.4246654510498047, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.7803, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 24.6468289750526, | |
| "eval_accuracy": 0.8195798531764643, | |
| "eval_loss": 1.0466234683990479, | |
| "eval_runtime": 114.7779, | |
| "eval_samples_per_second": 207.793, | |
| "eval_steps_per_second": 6.5, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 24.79711451758341, | |
| "grad_norm": 3.0004007816314697, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.7688, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 24.79711451758341, | |
| "eval_accuracy": 0.8197989075740256, | |
| "eval_loss": 1.0529950857162476, | |
| "eval_runtime": 113.2072, | |
| "eval_samples_per_second": 210.676, | |
| "eval_steps_per_second": 6.59, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 24.947400060114216, | |
| "grad_norm": 2.476900577545166, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.7734, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 24.947400060114216, | |
| "eval_accuracy": 0.8198411078292097, | |
| "eval_loss": 1.0492240190505981, | |
| "eval_runtime": 114.2362, | |
| "eval_samples_per_second": 208.778, | |
| "eval_steps_per_second": 6.53, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 25.097685602645026, | |
| "grad_norm": 2.6050217151641846, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.7741, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 25.097685602645026, | |
| "eval_accuracy": 0.8199209388676126, | |
| "eval_loss": 1.0443811416625977, | |
| "eval_runtime": 115.0076, | |
| "eval_samples_per_second": 207.378, | |
| "eval_steps_per_second": 6.487, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 25.247971145175836, | |
| "grad_norm": 2.845093011856079, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.768, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 25.247971145175836, | |
| "eval_accuracy": 0.8203943019525041, | |
| "eval_loss": 1.0556350946426392, | |
| "eval_runtime": 115.581, | |
| "eval_samples_per_second": 206.349, | |
| "eval_steps_per_second": 6.454, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 25.398256687706642, | |
| "grad_norm": 2.8715054988861084, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.7731, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 25.398256687706642, | |
| "eval_accuracy": 0.8204570484351956, | |
| "eval_loss": 1.0443503856658936, | |
| "eval_runtime": 114.2662, | |
| "eval_samples_per_second": 208.723, | |
| "eval_steps_per_second": 6.529, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 25.54854223023745, | |
| "grad_norm": 2.2550415992736816, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.7675, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 25.54854223023745, | |
| "eval_accuracy": 0.8200469636032075, | |
| "eval_loss": 1.0414886474609375, | |
| "eval_runtime": 114.9948, | |
| "eval_samples_per_second": 207.401, | |
| "eval_steps_per_second": 6.487, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 25.69882777276826, | |
| "grad_norm": 2.238607168197632, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.7675, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 25.69882777276826, | |
| "eval_accuracy": 0.8207652275282068, | |
| "eval_loss": 1.0444408655166626, | |
| "eval_runtime": 115.1961, | |
| "eval_samples_per_second": 207.038, | |
| "eval_steps_per_second": 6.476, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 25.849113315299068, | |
| "grad_norm": 2.242630958557129, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.7655, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 25.849113315299068, | |
| "eval_accuracy": 0.8202594142538011, | |
| "eval_loss": 1.0470616817474365, | |
| "eval_runtime": 116.2229, | |
| "eval_samples_per_second": 205.209, | |
| "eval_steps_per_second": 6.419, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 25.999398857829878, | |
| "grad_norm": 2.5453295707702637, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.7678, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 25.999398857829878, | |
| "eval_accuracy": 0.8200539868874701, | |
| "eval_loss": 1.0508933067321777, | |
| "eval_runtime": 114.7052, | |
| "eval_samples_per_second": 207.924, | |
| "eval_steps_per_second": 6.504, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 26.149684400360684, | |
| "grad_norm": 2.391352415084839, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.7653, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 26.149684400360684, | |
| "eval_accuracy": 0.8205590231388606, | |
| "eval_loss": 1.0484099388122559, | |
| "eval_runtime": 115.0871, | |
| "eval_samples_per_second": 207.234, | |
| "eval_steps_per_second": 6.482, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 26.299969942891494, | |
| "grad_norm": 2.5071208477020264, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.7679, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 26.299969942891494, | |
| "eval_accuracy": 0.8217573625905127, | |
| "eval_loss": 1.0400173664093018, | |
| "eval_runtime": 113.1376, | |
| "eval_samples_per_second": 210.805, | |
| "eval_steps_per_second": 6.594, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 26.450255485422304, | |
| "grad_norm": 2.5453133583068848, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.7656, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 26.450255485422304, | |
| "eval_accuracy": 0.8213452575807276, | |
| "eval_loss": 1.0516611337661743, | |
| "eval_runtime": 113.3943, | |
| "eval_samples_per_second": 210.328, | |
| "eval_steps_per_second": 6.579, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 26.60054102795311, | |
| "grad_norm": 2.509098529815674, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.7593, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 26.60054102795311, | |
| "eval_accuracy": 0.8213436048487595, | |
| "eval_loss": 1.0389631986618042, | |
| "eval_runtime": 116.4627, | |
| "eval_samples_per_second": 204.787, | |
| "eval_steps_per_second": 6.405, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 26.75082657048392, | |
| "grad_norm": 2.835665464401245, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.7606, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 26.75082657048392, | |
| "eval_accuracy": 0.8212840552795833, | |
| "eval_loss": 1.0403192043304443, | |
| "eval_runtime": 114.9351, | |
| "eval_samples_per_second": 207.508, | |
| "eval_steps_per_second": 6.491, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 26.90111211301473, | |
| "grad_norm": 2.5455708503723145, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.7578, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 26.90111211301473, | |
| "eval_accuracy": 0.821311120726958, | |
| "eval_loss": 1.036601185798645, | |
| "eval_runtime": 114.3227, | |
| "eval_samples_per_second": 208.62, | |
| "eval_steps_per_second": 6.525, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 27.051397655545536, | |
| "grad_norm": 2.821378231048584, | |
| "learning_rate": 1e-05, | |
| "loss": 0.765, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 27.051397655545536, | |
| "eval_accuracy": 0.8221304808698725, | |
| "eval_loss": 1.0333795547485352, | |
| "eval_runtime": 115.4024, | |
| "eval_samples_per_second": 206.668, | |
| "eval_steps_per_second": 6.464, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 27.201683198076346, | |
| "grad_norm": 2.59919810295105, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.7589, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 27.201683198076346, | |
| "eval_accuracy": 0.8222628454875403, | |
| "eval_loss": 1.033319354057312, | |
| "eval_runtime": 115.8055, | |
| "eval_samples_per_second": 205.949, | |
| "eval_steps_per_second": 6.442, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 27.351968740607152, | |
| "grad_norm": 2.601203203201294, | |
| "learning_rate": 9e-06, | |
| "loss": 0.7563, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 27.351968740607152, | |
| "eval_accuracy": 0.8218045068983223, | |
| "eval_loss": 1.046338677406311, | |
| "eval_runtime": 116.2837, | |
| "eval_samples_per_second": 205.102, | |
| "eval_steps_per_second": 6.415, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 27.50225428313796, | |
| "grad_norm": 2.8450610637664795, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.7559, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 27.50225428313796, | |
| "eval_accuracy": 0.8218716317162797, | |
| "eval_loss": 1.044384241104126, | |
| "eval_runtime": 115.6484, | |
| "eval_samples_per_second": 206.229, | |
| "eval_steps_per_second": 6.451, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 27.65253982566877, | |
| "grad_norm": 2.7283740043640137, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.7586, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 27.65253982566877, | |
| "eval_accuracy": 0.8221257250457026, | |
| "eval_loss": 1.0296134948730469, | |
| "eval_runtime": 115.214, | |
| "eval_samples_per_second": 207.006, | |
| "eval_steps_per_second": 6.475, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 27.802825368199578, | |
| "grad_norm": 2.795022964477539, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.7585, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 27.802825368199578, | |
| "eval_accuracy": 0.82162242799983, | |
| "eval_loss": 1.045117735862732, | |
| "eval_runtime": 113.2778, | |
| "eval_samples_per_second": 210.544, | |
| "eval_steps_per_second": 6.586, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 27.953110910730388, | |
| "grad_norm": 2.52536678314209, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.7548, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 27.953110910730388, | |
| "eval_accuracy": 0.8225299354468154, | |
| "eval_loss": 1.0380265712738037, | |
| "eval_runtime": 113.5711, | |
| "eval_samples_per_second": 210.001, | |
| "eval_steps_per_second": 6.569, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 28.103396453261198, | |
| "grad_norm": 2.3631057739257812, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.7542, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 28.103396453261198, | |
| "eval_accuracy": 0.8229969642015583, | |
| "eval_loss": 1.0490919351577759, | |
| "eval_runtime": 112.4838, | |
| "eval_samples_per_second": 212.031, | |
| "eval_steps_per_second": 6.632, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 28.253681995792004, | |
| "grad_norm": 2.2699172496795654, | |
| "learning_rate": 6e-06, | |
| "loss": 0.7522, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 28.253681995792004, | |
| "eval_accuracy": 0.8230864296098176, | |
| "eval_loss": 1.0274651050567627, | |
| "eval_runtime": 111.8716, | |
| "eval_samples_per_second": 213.191, | |
| "eval_steps_per_second": 6.668, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 28.403967538322814, | |
| "grad_norm": 2.44769287109375, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.7569, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 28.403967538322814, | |
| "eval_accuracy": 0.8234413637256175, | |
| "eval_loss": 1.0298686027526855, | |
| "eval_runtime": 113.5377, | |
| "eval_samples_per_second": 210.062, | |
| "eval_steps_per_second": 6.571, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 28.554253080853623, | |
| "grad_norm": 2.7103466987609863, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7536, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 28.554253080853623, | |
| "eval_accuracy": 0.8224712064035926, | |
| "eval_loss": 1.034175992012024, | |
| "eval_runtime": 113.1535, | |
| "eval_samples_per_second": 210.776, | |
| "eval_steps_per_second": 6.593, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 28.70453862338443, | |
| "grad_norm": 2.594381809234619, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.7547, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 28.70453862338443, | |
| "eval_accuracy": 0.8231576514254546, | |
| "eval_loss": 1.0324558019638062, | |
| "eval_runtime": 111.2443, | |
| "eval_samples_per_second": 214.393, | |
| "eval_steps_per_second": 6.706, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 28.85482416591524, | |
| "grad_norm": 2.623845100402832, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.7529, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 28.85482416591524, | |
| "eval_accuracy": 0.8241724336625226, | |
| "eval_loss": 1.0312702655792236, | |
| "eval_runtime": 115.266, | |
| "eval_samples_per_second": 206.913, | |
| "eval_steps_per_second": 6.472, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 29.005109708446046, | |
| "grad_norm": 2.706256628036499, | |
| "learning_rate": 3.5000000000000004e-06, | |
| "loss": 0.7461, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 29.005109708446046, | |
| "eval_accuracy": 0.8230462104005479, | |
| "eval_loss": 1.0389125347137451, | |
| "eval_runtime": 115.814, | |
| "eval_samples_per_second": 205.934, | |
| "eval_steps_per_second": 6.441, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 29.155395250976856, | |
| "grad_norm": 2.697993278503418, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7513, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 29.155395250976856, | |
| "eval_accuracy": 0.8236184494498082, | |
| "eval_loss": 1.036423921585083, | |
| "eval_runtime": 115.3283, | |
| "eval_samples_per_second": 206.801, | |
| "eval_steps_per_second": 6.468, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 29.305680793507666, | |
| "grad_norm": 2.68867826461792, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.7494, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 29.305680793507666, | |
| "eval_accuracy": 0.82393844898435, | |
| "eval_loss": 1.0242578983306885, | |
| "eval_runtime": 116.3698, | |
| "eval_samples_per_second": 204.95, | |
| "eval_steps_per_second": 6.411, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 29.455966336038472, | |
| "grad_norm": 2.259347915649414, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.7438, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 29.455966336038472, | |
| "eval_accuracy": 0.8236859969210806, | |
| "eval_loss": 1.0227982997894287, | |
| "eval_runtime": 115.3714, | |
| "eval_samples_per_second": 206.724, | |
| "eval_steps_per_second": 6.466, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 29.60625187856928, | |
| "grad_norm": 2.8278329372406006, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.7499, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 29.60625187856928, | |
| "eval_accuracy": 0.8236034478162941, | |
| "eval_loss": 1.022267460823059, | |
| "eval_runtime": 114.5023, | |
| "eval_samples_per_second": 208.293, | |
| "eval_steps_per_second": 6.515, | |
| "step": 98500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 31, | |
| "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": 8.31746763541971e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |