| { | |
| "best_metric": 0.8419870138168335, | |
| "best_model_checkpoint": "./model_fine-tune/glot/xlm-r/amh-Ethi/checkpoint-94500", | |
| "epoch": 121.46529562982005, | |
| "eval_steps": 500, | |
| "global_step": 94500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6426735218508998, | |
| "grad_norm": 4.40508508682251, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.6805, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6426735218508998, | |
| "eval_accuracy": 0.6904924931069853, | |
| "eval_loss": 1.4787547588348389, | |
| "eval_runtime": 34.6674, | |
| "eval_samples_per_second": 179.506, | |
| "eval_steps_per_second": 5.625, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2853470437017995, | |
| "grad_norm": 3.9189295768737793, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.5356, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.2853470437017995, | |
| "eval_accuracy": 0.7092839911070465, | |
| "eval_loss": 1.3684136867523193, | |
| "eval_runtime": 35.1037, | |
| "eval_samples_per_second": 177.275, | |
| "eval_steps_per_second": 5.555, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.9280205655526992, | |
| "grad_norm": 4.308630466461182, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.4583, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.9280205655526992, | |
| "eval_accuracy": 0.7174822425497339, | |
| "eval_loss": 1.3316432237625122, | |
| "eval_runtime": 32.7687, | |
| "eval_samples_per_second": 189.907, | |
| "eval_steps_per_second": 5.951, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.570694087403599, | |
| "grad_norm": 3.4791815280914307, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.4029, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.570694087403599, | |
| "eval_accuracy": 0.7272827658735376, | |
| "eval_loss": 1.280930519104004, | |
| "eval_runtime": 35.7018, | |
| "eval_samples_per_second": 174.305, | |
| "eval_steps_per_second": 5.462, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.2133676092544987, | |
| "grad_norm": 3.705988645553589, | |
| "learning_rate": 9.75e-05, | |
| "loss": 1.3485, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.2133676092544987, | |
| "eval_accuracy": 0.7312191932269085, | |
| "eval_loss": 1.2635687589645386, | |
| "eval_runtime": 34.9687, | |
| "eval_samples_per_second": 177.959, | |
| "eval_steps_per_second": 5.576, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.8560411311053984, | |
| "grad_norm": 3.63777232170105, | |
| "learning_rate": 9.7e-05, | |
| "loss": 1.318, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.8560411311053984, | |
| "eval_accuracy": 0.7356960949650779, | |
| "eval_loss": 1.2287126779556274, | |
| "eval_runtime": 35.1013, | |
| "eval_samples_per_second": 177.287, | |
| "eval_steps_per_second": 5.555, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.4987146529562985, | |
| "grad_norm": 3.065805673599243, | |
| "learning_rate": 9.65e-05, | |
| "loss": 1.2853, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.4987146529562985, | |
| "eval_accuracy": 0.7411218708879311, | |
| "eval_loss": 1.2096679210662842, | |
| "eval_runtime": 35.3152, | |
| "eval_samples_per_second": 176.213, | |
| "eval_steps_per_second": 5.522, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.141388174807198, | |
| "grad_norm": 3.083160161972046, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.2579, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.141388174807198, | |
| "eval_accuracy": 0.743514192750392, | |
| "eval_loss": 1.1946351528167725, | |
| "eval_runtime": 35.3519, | |
| "eval_samples_per_second": 176.03, | |
| "eval_steps_per_second": 5.516, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.784061696658098, | |
| "grad_norm": 3.1521048545837402, | |
| "learning_rate": 9.55e-05, | |
| "loss": 1.225, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.784061696658098, | |
| "eval_accuracy": 0.7478054754473986, | |
| "eval_loss": 1.1656205654144287, | |
| "eval_runtime": 35.2967, | |
| "eval_samples_per_second": 176.305, | |
| "eval_steps_per_second": 5.525, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.426735218508997, | |
| "grad_norm": 2.947038412094116, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.2156, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.426735218508997, | |
| "eval_accuracy": 0.7494688799077729, | |
| "eval_loss": 1.1547260284423828, | |
| "eval_runtime": 35.244, | |
| "eval_samples_per_second": 176.569, | |
| "eval_steps_per_second": 5.533, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.069408740359897, | |
| "grad_norm": 3.245880603790283, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 1.1878, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.069408740359897, | |
| "eval_accuracy": 0.7533389855345517, | |
| "eval_loss": 1.1394718885421753, | |
| "eval_runtime": 34.1955, | |
| "eval_samples_per_second": 181.983, | |
| "eval_steps_per_second": 5.703, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 7.712082262210797, | |
| "grad_norm": 3.02457857131958, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.1713, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.712082262210797, | |
| "eval_accuracy": 0.7548033682748899, | |
| "eval_loss": 1.136972188949585, | |
| "eval_runtime": 33.9827, | |
| "eval_samples_per_second": 183.123, | |
| "eval_steps_per_second": 5.738, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.354755784061696, | |
| "grad_norm": 2.8089816570281982, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 1.1587, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.354755784061696, | |
| "eval_accuracy": 0.7558409318256807, | |
| "eval_loss": 1.1281076669692993, | |
| "eval_runtime": 34.9074, | |
| "eval_samples_per_second": 178.272, | |
| "eval_steps_per_second": 5.586, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.997429305912597, | |
| "grad_norm": 2.808964729309082, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 1.1416, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.997429305912597, | |
| "eval_accuracy": 0.7600179626094105, | |
| "eval_loss": 1.107399821281433, | |
| "eval_runtime": 34.0824, | |
| "eval_samples_per_second": 182.587, | |
| "eval_steps_per_second": 5.721, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 9.640102827763496, | |
| "grad_norm": 2.8558380603790283, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 1.128, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.640102827763496, | |
| "eval_accuracy": 0.7628805020537848, | |
| "eval_loss": 1.0930116176605225, | |
| "eval_runtime": 33.4368, | |
| "eval_samples_per_second": 186.112, | |
| "eval_steps_per_second": 5.832, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 10.282776349614396, | |
| "grad_norm": 2.9800262451171875, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 1.1213, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.282776349614396, | |
| "eval_accuracy": 0.7627036601926251, | |
| "eval_loss": 1.0944606065750122, | |
| "eval_runtime": 34.4461, | |
| "eval_samples_per_second": 180.659, | |
| "eval_steps_per_second": 5.661, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 10.925449871465295, | |
| "grad_norm": 3.040515422821045, | |
| "learning_rate": 9.15e-05, | |
| "loss": 1.1075, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 10.925449871465295, | |
| "eval_accuracy": 0.764068771058977, | |
| "eval_loss": 1.0873337984085083, | |
| "eval_runtime": 32.7071, | |
| "eval_samples_per_second": 190.265, | |
| "eval_steps_per_second": 5.962, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 11.568123393316196, | |
| "grad_norm": 2.973078966140747, | |
| "learning_rate": 9.1e-05, | |
| "loss": 1.0912, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.568123393316196, | |
| "eval_accuracy": 0.7658058887449325, | |
| "eval_loss": 1.0768815279006958, | |
| "eval_runtime": 32.6454, | |
| "eval_samples_per_second": 190.624, | |
| "eval_steps_per_second": 5.973, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 12.210796915167094, | |
| "grad_norm": 3.5373220443725586, | |
| "learning_rate": 9.05e-05, | |
| "loss": 1.085, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.210796915167094, | |
| "eval_accuracy": 0.7658247272593731, | |
| "eval_loss": 1.086098074913025, | |
| "eval_runtime": 32.662, | |
| "eval_samples_per_second": 190.527, | |
| "eval_steps_per_second": 5.97, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 12.853470437017995, | |
| "grad_norm": 5.5106964111328125, | |
| "learning_rate": 9e-05, | |
| "loss": 1.0746, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.853470437017995, | |
| "eval_accuracy": 0.7682209938962093, | |
| "eval_loss": 1.0635614395141602, | |
| "eval_runtime": 32.6593, | |
| "eval_samples_per_second": 190.543, | |
| "eval_steps_per_second": 5.971, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 13.496143958868895, | |
| "grad_norm": 3.238605499267578, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 1.0531, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 13.496143958868895, | |
| "eval_accuracy": 0.7686876543500863, | |
| "eval_loss": 1.0645390748977661, | |
| "eval_runtime": 32.6258, | |
| "eval_samples_per_second": 190.739, | |
| "eval_steps_per_second": 5.977, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 14.138817480719794, | |
| "grad_norm": 2.7707929611206055, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 1.0545, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.138817480719794, | |
| "eval_accuracy": 0.7713913141190644, | |
| "eval_loss": 1.0523449182510376, | |
| "eval_runtime": 33.0121, | |
| "eval_samples_per_second": 188.507, | |
| "eval_steps_per_second": 5.907, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 14.781491002570695, | |
| "grad_norm": 3.1346287727355957, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 1.042, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 14.781491002570695, | |
| "eval_accuracy": 0.7711833853943258, | |
| "eval_loss": 1.0591264963150024, | |
| "eval_runtime": 32.5788, | |
| "eval_samples_per_second": 191.014, | |
| "eval_steps_per_second": 5.985, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 15.424164524421593, | |
| "grad_norm": 2.98738169670105, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.0327, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 15.424164524421593, | |
| "eval_accuracy": 0.7727139790786457, | |
| "eval_loss": 1.0433602333068848, | |
| "eval_runtime": 32.6912, | |
| "eval_samples_per_second": 190.357, | |
| "eval_steps_per_second": 5.965, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 16.066838046272494, | |
| "grad_norm": 2.7987964153289795, | |
| "learning_rate": 8.75e-05, | |
| "loss": 1.0226, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 16.066838046272494, | |
| "eval_accuracy": 0.7744916530689092, | |
| "eval_loss": 1.0427190065383911, | |
| "eval_runtime": 32.7105, | |
| "eval_samples_per_second": 190.245, | |
| "eval_steps_per_second": 5.961, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 16.709511568123393, | |
| "grad_norm": 2.927727699279785, | |
| "learning_rate": 8.7e-05, | |
| "loss": 1.0164, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 16.709511568123393, | |
| "eval_accuracy": 0.7755782307589745, | |
| "eval_loss": 1.033629059791565, | |
| "eval_runtime": 32.6997, | |
| "eval_samples_per_second": 190.307, | |
| "eval_steps_per_second": 5.963, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 17.35218508997429, | |
| "grad_norm": 2.902130603790283, | |
| "learning_rate": 8.65e-05, | |
| "loss": 1.012, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 17.35218508997429, | |
| "eval_accuracy": 0.7768737795368466, | |
| "eval_loss": 1.024613380432129, | |
| "eval_runtime": 32.6543, | |
| "eval_samples_per_second": 190.572, | |
| "eval_steps_per_second": 5.972, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 17.994858611825194, | |
| "grad_norm": 2.6647133827209473, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.9992, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 17.994858611825194, | |
| "eval_accuracy": 0.7779743325654593, | |
| "eval_loss": 1.0130186080932617, | |
| "eval_runtime": 32.7375, | |
| "eval_samples_per_second": 190.088, | |
| "eval_steps_per_second": 5.956, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 18.637532133676093, | |
| "grad_norm": 2.590914249420166, | |
| "learning_rate": 8.55e-05, | |
| "loss": 0.9928, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 18.637532133676093, | |
| "eval_accuracy": 0.7779691811484551, | |
| "eval_loss": 1.0204987525939941, | |
| "eval_runtime": 33.4597, | |
| "eval_samples_per_second": 185.985, | |
| "eval_steps_per_second": 5.828, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 19.28020565552699, | |
| "grad_norm": 2.6548004150390625, | |
| "learning_rate": 8.5e-05, | |
| "loss": 0.985, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 19.28020565552699, | |
| "eval_accuracy": 0.7788039344262295, | |
| "eval_loss": 1.0153322219848633, | |
| "eval_runtime": 34.4688, | |
| "eval_samples_per_second": 180.54, | |
| "eval_steps_per_second": 5.657, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 19.922879177377894, | |
| "grad_norm": 2.6247718334198, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 0.9822, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 19.922879177377894, | |
| "eval_accuracy": 0.7795580110497238, | |
| "eval_loss": 1.0121690034866333, | |
| "eval_runtime": 34.3855, | |
| "eval_samples_per_second": 180.977, | |
| "eval_steps_per_second": 5.671, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 20.565552699228792, | |
| "grad_norm": 2.7304604053497314, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.9735, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 20.565552699228792, | |
| "eval_accuracy": 0.7793682688302468, | |
| "eval_loss": 1.018441081047058, | |
| "eval_runtime": 34.5499, | |
| "eval_samples_per_second": 180.116, | |
| "eval_steps_per_second": 5.644, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 21.20822622107969, | |
| "grad_norm": 2.7982735633850098, | |
| "learning_rate": 8.35e-05, | |
| "loss": 0.971, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 21.20822622107969, | |
| "eval_accuracy": 0.7816012690588251, | |
| "eval_loss": 1.0020060539245605, | |
| "eval_runtime": 34.6648, | |
| "eval_samples_per_second": 179.519, | |
| "eval_steps_per_second": 5.625, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 21.85089974293059, | |
| "grad_norm": 2.6124517917633057, | |
| "learning_rate": 8.3e-05, | |
| "loss": 0.965, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 21.85089974293059, | |
| "eval_accuracy": 0.7828390025055424, | |
| "eval_loss": 0.9919770359992981, | |
| "eval_runtime": 34.2625, | |
| "eval_samples_per_second": 181.627, | |
| "eval_steps_per_second": 5.691, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 22.493573264781492, | |
| "grad_norm": 2.7260468006134033, | |
| "learning_rate": 8.25e-05, | |
| "loss": 0.9536, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 22.493573264781492, | |
| "eval_accuracy": 0.7826561078737454, | |
| "eval_loss": 0.9964642524719238, | |
| "eval_runtime": 35.1121, | |
| "eval_samples_per_second": 177.232, | |
| "eval_steps_per_second": 5.554, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 23.13624678663239, | |
| "grad_norm": 2.5959136486053467, | |
| "learning_rate": 8.2e-05, | |
| "loss": 0.9515, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 23.13624678663239, | |
| "eval_accuracy": 0.7842808178702281, | |
| "eval_loss": 0.9843628406524658, | |
| "eval_runtime": 35.1683, | |
| "eval_samples_per_second": 176.949, | |
| "eval_steps_per_second": 5.545, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 23.77892030848329, | |
| "grad_norm": 2.8720030784606934, | |
| "learning_rate": 8.15e-05, | |
| "loss": 0.9459, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 23.77892030848329, | |
| "eval_accuracy": 0.7822331893892659, | |
| "eval_loss": 0.9950876235961914, | |
| "eval_runtime": 35.4899, | |
| "eval_samples_per_second": 175.345, | |
| "eval_steps_per_second": 5.495, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 24.42159383033419, | |
| "grad_norm": 3.0184552669525146, | |
| "learning_rate": 8.1e-05, | |
| "loss": 0.9374, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 24.42159383033419, | |
| "eval_accuracy": 0.7852158669501615, | |
| "eval_loss": 0.9859277009963989, | |
| "eval_runtime": 34.116, | |
| "eval_samples_per_second": 182.407, | |
| "eval_steps_per_second": 5.716, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 25.06426735218509, | |
| "grad_norm": 2.516874074935913, | |
| "learning_rate": 8.05e-05, | |
| "loss": 0.9318, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 25.06426735218509, | |
| "eval_accuracy": 0.7829394214826243, | |
| "eval_loss": 0.9957559108734131, | |
| "eval_runtime": 35.0753, | |
| "eval_samples_per_second": 177.419, | |
| "eval_steps_per_second": 5.559, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 25.70694087403599, | |
| "grad_norm": 2.5600101947784424, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9311, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 25.70694087403599, | |
| "eval_accuracy": 0.786506985609304, | |
| "eval_loss": 0.9760250449180603, | |
| "eval_runtime": 34.6952, | |
| "eval_samples_per_second": 179.362, | |
| "eval_steps_per_second": 5.62, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 26.34961439588689, | |
| "grad_norm": 2.7094664573669434, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 0.913, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 26.34961439588689, | |
| "eval_accuracy": 0.78658190904128, | |
| "eval_loss": 0.9806557297706604, | |
| "eval_runtime": 34.3785, | |
| "eval_samples_per_second": 181.014, | |
| "eval_steps_per_second": 5.672, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 26.99228791773779, | |
| "grad_norm": 2.3693931102752686, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 0.9171, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 26.99228791773779, | |
| "eval_accuracy": 0.7854132867260752, | |
| "eval_loss": 0.9825689196586609, | |
| "eval_runtime": 34.4341, | |
| "eval_samples_per_second": 180.722, | |
| "eval_steps_per_second": 5.663, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 27.63496143958869, | |
| "grad_norm": 3.0254580974578857, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 0.9062, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 27.63496143958869, | |
| "eval_accuracy": 0.7879520726883571, | |
| "eval_loss": 0.9732692241668701, | |
| "eval_runtime": 34.5783, | |
| "eval_samples_per_second": 179.968, | |
| "eval_steps_per_second": 5.639, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 28.27763496143959, | |
| "grad_norm": 2.6540822982788086, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.9095, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 28.27763496143959, | |
| "eval_accuracy": 0.7879547877788341, | |
| "eval_loss": 0.9751275777816772, | |
| "eval_runtime": 33.0181, | |
| "eval_samples_per_second": 188.473, | |
| "eval_steps_per_second": 5.906, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 28.920308483290487, | |
| "grad_norm": 2.7978150844573975, | |
| "learning_rate": 7.75e-05, | |
| "loss": 0.9003, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 28.920308483290487, | |
| "eval_accuracy": 0.789157797413295, | |
| "eval_loss": 0.9676902890205383, | |
| "eval_runtime": 34.176, | |
| "eval_samples_per_second": 182.087, | |
| "eval_steps_per_second": 5.706, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 29.56298200514139, | |
| "grad_norm": 2.6367812156677246, | |
| "learning_rate": 7.7e-05, | |
| "loss": 0.8994, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 29.56298200514139, | |
| "eval_accuracy": 0.78875261302136, | |
| "eval_loss": 0.9720832109451294, | |
| "eval_runtime": 34.6435, | |
| "eval_samples_per_second": 179.63, | |
| "eval_steps_per_second": 5.629, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 30.205655526992288, | |
| "grad_norm": 2.8410837650299072, | |
| "learning_rate": 7.65e-05, | |
| "loss": 0.8937, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 30.205655526992288, | |
| "eval_accuracy": 0.7898726325583352, | |
| "eval_loss": 0.9666265249252319, | |
| "eval_runtime": 34.382, | |
| "eval_samples_per_second": 180.996, | |
| "eval_steps_per_second": 5.672, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 30.848329048843187, | |
| "grad_norm": 2.6249313354492188, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.8853, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 30.848329048843187, | |
| "eval_accuracy": 0.7914164731021732, | |
| "eval_loss": 0.9592417478561401, | |
| "eval_runtime": 34.7355, | |
| "eval_samples_per_second": 179.154, | |
| "eval_steps_per_second": 5.614, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 31.491002570694086, | |
| "grad_norm": 2.563669443130493, | |
| "learning_rate": 7.55e-05, | |
| "loss": 0.8838, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 31.491002570694086, | |
| "eval_accuracy": 0.7892845518211694, | |
| "eval_loss": 0.969780683517456, | |
| "eval_runtime": 34.7542, | |
| "eval_samples_per_second": 179.057, | |
| "eval_steps_per_second": 5.611, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 32.13367609254499, | |
| "grad_norm": 2.7462027072906494, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.8809, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 32.13367609254499, | |
| "eval_accuracy": 0.7920457676713335, | |
| "eval_loss": 0.9529361128807068, | |
| "eval_runtime": 34.3098, | |
| "eval_samples_per_second": 181.377, | |
| "eval_steps_per_second": 5.684, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 32.77634961439589, | |
| "grad_norm": 2.533267021179199, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 0.874, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 32.77634961439589, | |
| "eval_accuracy": 0.7919883512638669, | |
| "eval_loss": 0.9449800252914429, | |
| "eval_runtime": 34.7988, | |
| "eval_samples_per_second": 178.828, | |
| "eval_steps_per_second": 5.604, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 33.419023136246786, | |
| "grad_norm": 2.718332529067993, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.8647, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 33.419023136246786, | |
| "eval_accuracy": 0.7927094725330193, | |
| "eval_loss": 0.9489922523498535, | |
| "eval_runtime": 34.773, | |
| "eval_samples_per_second": 178.961, | |
| "eval_steps_per_second": 5.608, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 34.06169665809769, | |
| "grad_norm": 2.664407253265381, | |
| "learning_rate": 7.35e-05, | |
| "loss": 0.8638, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 34.06169665809769, | |
| "eval_accuracy": 0.7909423101119826, | |
| "eval_loss": 0.9638357758522034, | |
| "eval_runtime": 34.4764, | |
| "eval_samples_per_second": 180.501, | |
| "eval_steps_per_second": 5.656, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 34.70437017994858, | |
| "grad_norm": 2.9981689453125, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.8604, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 34.70437017994858, | |
| "eval_accuracy": 0.792258094990671, | |
| "eval_loss": 0.9567773938179016, | |
| "eval_runtime": 34.4124, | |
| "eval_samples_per_second": 180.836, | |
| "eval_steps_per_second": 5.667, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 35.347043701799485, | |
| "grad_norm": 2.6397457122802734, | |
| "learning_rate": 7.25e-05, | |
| "loss": 0.8516, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 35.347043701799485, | |
| "eval_accuracy": 0.7944613334662306, | |
| "eval_loss": 0.9480540156364441, | |
| "eval_runtime": 34.3734, | |
| "eval_samples_per_second": 181.041, | |
| "eval_steps_per_second": 5.673, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 35.98971722365039, | |
| "grad_norm": 2.9263813495635986, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.8529, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 35.98971722365039, | |
| "eval_accuracy": 0.7949814058148135, | |
| "eval_loss": 0.9380202293395996, | |
| "eval_runtime": 34.1807, | |
| "eval_samples_per_second": 182.062, | |
| "eval_steps_per_second": 5.705, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 36.63239074550128, | |
| "grad_norm": 2.446913719177246, | |
| "learning_rate": 7.15e-05, | |
| "loss": 0.8454, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 36.63239074550128, | |
| "eval_accuracy": 0.7940900415285876, | |
| "eval_loss": 0.9452695250511169, | |
| "eval_runtime": 33.1944, | |
| "eval_samples_per_second": 187.472, | |
| "eval_steps_per_second": 5.874, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 37.275064267352185, | |
| "grad_norm": 2.429306983947754, | |
| "learning_rate": 7.1e-05, | |
| "loss": 0.8432, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 37.275064267352185, | |
| "eval_accuracy": 0.7944350218702906, | |
| "eval_loss": 0.9466774463653564, | |
| "eval_runtime": 33.291, | |
| "eval_samples_per_second": 186.927, | |
| "eval_steps_per_second": 5.857, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 37.91773778920309, | |
| "grad_norm": 2.648399591445923, | |
| "learning_rate": 7.05e-05, | |
| "loss": 0.8468, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 37.91773778920309, | |
| "eval_accuracy": 0.7957481551099058, | |
| "eval_loss": 0.9400731325149536, | |
| "eval_runtime": 33.2405, | |
| "eval_samples_per_second": 187.211, | |
| "eval_steps_per_second": 5.866, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 38.56041131105398, | |
| "grad_norm": 2.644608974456787, | |
| "learning_rate": 7e-05, | |
| "loss": 0.8328, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 38.56041131105398, | |
| "eval_accuracy": 0.795926071569013, | |
| "eval_loss": 0.9422577619552612, | |
| "eval_runtime": 33.3493, | |
| "eval_samples_per_second": 186.601, | |
| "eval_steps_per_second": 5.847, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 39.203084832904885, | |
| "grad_norm": 2.6563777923583984, | |
| "learning_rate": 6.95e-05, | |
| "loss": 0.8323, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 39.203084832904885, | |
| "eval_accuracy": 0.7951947535355016, | |
| "eval_loss": 0.9368035793304443, | |
| "eval_runtime": 32.9628, | |
| "eval_samples_per_second": 188.788, | |
| "eval_steps_per_second": 5.916, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 39.84575835475579, | |
| "grad_norm": 2.5932886600494385, | |
| "learning_rate": 6.9e-05, | |
| "loss": 0.8268, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 39.84575835475579, | |
| "eval_accuracy": 0.7967557476448579, | |
| "eval_loss": 0.9303238391876221, | |
| "eval_runtime": 32.4851, | |
| "eval_samples_per_second": 191.565, | |
| "eval_steps_per_second": 6.003, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 40.48843187660668, | |
| "grad_norm": 2.7462244033813477, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 0.8246, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 40.48843187660668, | |
| "eval_accuracy": 0.7966305277950054, | |
| "eval_loss": 0.9323325157165527, | |
| "eval_runtime": 33.1165, | |
| "eval_samples_per_second": 187.912, | |
| "eval_steps_per_second": 5.888, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 41.131105398457585, | |
| "grad_norm": 2.647944211959839, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.8216, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 41.131105398457585, | |
| "eval_accuracy": 0.7962699125415105, | |
| "eval_loss": 0.9350172281265259, | |
| "eval_runtime": 32.873, | |
| "eval_samples_per_second": 189.305, | |
| "eval_steps_per_second": 5.932, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 41.77377892030848, | |
| "grad_norm": 2.661036729812622, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 0.8218, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 41.77377892030848, | |
| "eval_accuracy": 0.7972857023271388, | |
| "eval_loss": 0.9303329586982727, | |
| "eval_runtime": 32.6506, | |
| "eval_samples_per_second": 190.594, | |
| "eval_steps_per_second": 5.972, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 42.41645244215938, | |
| "grad_norm": 2.7694153785705566, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.8132, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 42.41645244215938, | |
| "eval_accuracy": 0.7977203787112787, | |
| "eval_loss": 0.9363131523132324, | |
| "eval_runtime": 33.0889, | |
| "eval_samples_per_second": 188.069, | |
| "eval_steps_per_second": 5.893, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 43.059125964010285, | |
| "grad_norm": 2.8761026859283447, | |
| "learning_rate": 6.65e-05, | |
| "loss": 0.8082, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 43.059125964010285, | |
| "eval_accuracy": 0.7981091852326369, | |
| "eval_loss": 0.9364966154098511, | |
| "eval_runtime": 32.3923, | |
| "eval_samples_per_second": 192.113, | |
| "eval_steps_per_second": 6.02, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 43.70179948586118, | |
| "grad_norm": 2.6908044815063477, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.806, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 43.70179948586118, | |
| "eval_accuracy": 0.7986442847736496, | |
| "eval_loss": 0.9224198460578918, | |
| "eval_runtime": 33.388, | |
| "eval_samples_per_second": 186.384, | |
| "eval_steps_per_second": 5.84, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 44.34447300771208, | |
| "grad_norm": 2.6623470783233643, | |
| "learning_rate": 6.55e-05, | |
| "loss": 0.8024, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 44.34447300771208, | |
| "eval_accuracy": 0.7992224783044672, | |
| "eval_loss": 0.9192689061164856, | |
| "eval_runtime": 33.3722, | |
| "eval_samples_per_second": 186.473, | |
| "eval_steps_per_second": 5.843, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 44.987146529562985, | |
| "grad_norm": 2.6698567867279053, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.8016, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 44.987146529562985, | |
| "eval_accuracy": 0.7984851112126992, | |
| "eval_loss": 0.9178963303565979, | |
| "eval_runtime": 33.2663, | |
| "eval_samples_per_second": 187.066, | |
| "eval_steps_per_second": 5.862, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 45.62982005141388, | |
| "grad_norm": 2.980325222015381, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 0.796, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 45.62982005141388, | |
| "eval_accuracy": 0.7979037090039865, | |
| "eval_loss": 0.9273726344108582, | |
| "eval_runtime": 33.0665, | |
| "eval_samples_per_second": 188.196, | |
| "eval_steps_per_second": 5.897, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 46.27249357326478, | |
| "grad_norm": 2.536480188369751, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.7976, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 46.27249357326478, | |
| "eval_accuracy": 0.8009755918651776, | |
| "eval_loss": 0.9136722087860107, | |
| "eval_runtime": 33.7816, | |
| "eval_samples_per_second": 184.213, | |
| "eval_steps_per_second": 5.772, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 46.915167095115685, | |
| "grad_norm": 2.369147777557373, | |
| "learning_rate": 6.35e-05, | |
| "loss": 0.7888, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 46.915167095115685, | |
| "eval_accuracy": 0.8006721188810271, | |
| "eval_loss": 0.9139747619628906, | |
| "eval_runtime": 33.0443, | |
| "eval_samples_per_second": 188.323, | |
| "eval_steps_per_second": 5.901, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 47.55784061696658, | |
| "grad_norm": 2.606424570083618, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.7826, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 47.55784061696658, | |
| "eval_accuracy": 0.8006546893234877, | |
| "eval_loss": 0.9165197610855103, | |
| "eval_runtime": 32.9373, | |
| "eval_samples_per_second": 188.935, | |
| "eval_steps_per_second": 5.92, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 48.20051413881748, | |
| "grad_norm": 2.869199514389038, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.789, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 48.20051413881748, | |
| "eval_accuracy": 0.8009030033534118, | |
| "eval_loss": 0.9167375564575195, | |
| "eval_runtime": 33.2315, | |
| "eval_samples_per_second": 187.262, | |
| "eval_steps_per_second": 5.868, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 48.84318766066838, | |
| "grad_norm": 2.5320894718170166, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.7827, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 48.84318766066838, | |
| "eval_accuracy": 0.7997841239160237, | |
| "eval_loss": 0.9282008409500122, | |
| "eval_runtime": 33.3582, | |
| "eval_samples_per_second": 186.551, | |
| "eval_steps_per_second": 5.846, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 49.48586118251928, | |
| "grad_norm": 2.741992712020874, | |
| "learning_rate": 6.15e-05, | |
| "loss": 0.7739, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 49.48586118251928, | |
| "eval_accuracy": 0.8020669565628894, | |
| "eval_loss": 0.9078426361083984, | |
| "eval_runtime": 33.2299, | |
| "eval_samples_per_second": 187.271, | |
| "eval_steps_per_second": 5.868, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 50.12853470437018, | |
| "grad_norm": 2.563089370727539, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.7766, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 50.12853470437018, | |
| "eval_accuracy": 0.8028247073903633, | |
| "eval_loss": 0.9107823967933655, | |
| "eval_runtime": 33.5047, | |
| "eval_samples_per_second": 185.735, | |
| "eval_steps_per_second": 5.82, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 50.77120822622108, | |
| "grad_norm": 2.5732595920562744, | |
| "learning_rate": 6.05e-05, | |
| "loss": 0.7728, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 50.77120822622108, | |
| "eval_accuracy": 0.8025567474818919, | |
| "eval_loss": 0.9065195918083191, | |
| "eval_runtime": 33.1402, | |
| "eval_samples_per_second": 187.778, | |
| "eval_steps_per_second": 5.884, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 51.41388174807198, | |
| "grad_norm": 2.4717209339141846, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7676, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 51.41388174807198, | |
| "eval_accuracy": 0.8028714592601541, | |
| "eval_loss": 0.9060749411582947, | |
| "eval_runtime": 32.3856, | |
| "eval_samples_per_second": 192.153, | |
| "eval_steps_per_second": 6.021, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 52.05655526992288, | |
| "grad_norm": 2.5366320610046387, | |
| "learning_rate": 5.95e-05, | |
| "loss": 0.7653, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 52.05655526992288, | |
| "eval_accuracy": 0.8030310980056771, | |
| "eval_loss": 0.9148956537246704, | |
| "eval_runtime": 33.3546, | |
| "eval_samples_per_second": 186.571, | |
| "eval_steps_per_second": 5.846, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 52.69922879177378, | |
| "grad_norm": 2.9074325561523438, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.7607, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 52.69922879177378, | |
| "eval_accuracy": 0.8039304797381431, | |
| "eval_loss": 0.8974832892417908, | |
| "eval_runtime": 33.4483, | |
| "eval_samples_per_second": 186.049, | |
| "eval_steps_per_second": 5.83, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 53.34190231362468, | |
| "grad_norm": 3.1211259365081787, | |
| "learning_rate": 5.85e-05, | |
| "loss": 0.7604, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 53.34190231362468, | |
| "eval_accuracy": 0.8036355728594897, | |
| "eval_loss": 0.90870201587677, | |
| "eval_runtime": 33.4646, | |
| "eval_samples_per_second": 185.957, | |
| "eval_steps_per_second": 5.827, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 53.98457583547558, | |
| "grad_norm": 2.762848377227783, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.7593, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 53.98457583547558, | |
| "eval_accuracy": 0.802461854621613, | |
| "eval_loss": 0.9101512432098389, | |
| "eval_runtime": 32.3612, | |
| "eval_samples_per_second": 192.298, | |
| "eval_steps_per_second": 6.026, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 54.62724935732648, | |
| "grad_norm": 2.6548171043395996, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.7564, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 54.62724935732648, | |
| "eval_accuracy": 0.804102518442478, | |
| "eval_loss": 0.9050089120864868, | |
| "eval_runtime": 33.1101, | |
| "eval_samples_per_second": 187.949, | |
| "eval_steps_per_second": 5.889, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 55.26992287917738, | |
| "grad_norm": 2.6228950023651123, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.7452, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 55.26992287917738, | |
| "eval_accuracy": 0.8040075271997299, | |
| "eval_loss": 0.9055400490760803, | |
| "eval_runtime": 33.014, | |
| "eval_samples_per_second": 188.496, | |
| "eval_steps_per_second": 5.907, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 55.912596401028274, | |
| "grad_norm": 2.6490426063537598, | |
| "learning_rate": 5.65e-05, | |
| "loss": 0.7488, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 55.912596401028274, | |
| "eval_accuracy": 0.8048906062428912, | |
| "eval_loss": 0.9053667187690735, | |
| "eval_runtime": 32.3856, | |
| "eval_samples_per_second": 192.153, | |
| "eval_steps_per_second": 6.021, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 56.55526992287918, | |
| "grad_norm": 2.494063377380371, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.7472, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 56.55526992287918, | |
| "eval_accuracy": 0.8050882383051106, | |
| "eval_loss": 0.8990674614906311, | |
| "eval_runtime": 34.0904, | |
| "eval_samples_per_second": 182.544, | |
| "eval_steps_per_second": 5.72, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 57.19794344473008, | |
| "grad_norm": 2.8385369777679443, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 0.7469, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 57.19794344473008, | |
| "eval_accuracy": 0.8047514040498369, | |
| "eval_loss": 0.9029610753059387, | |
| "eval_runtime": 32.596, | |
| "eval_samples_per_second": 190.913, | |
| "eval_steps_per_second": 5.982, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 57.840616966580974, | |
| "grad_norm": 2.4786102771759033, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.7386, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 57.840616966580974, | |
| "eval_accuracy": 0.8045387474256496, | |
| "eval_loss": 0.8992837071418762, | |
| "eval_runtime": 33.7728, | |
| "eval_samples_per_second": 184.261, | |
| "eval_steps_per_second": 5.774, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 58.48329048843188, | |
| "grad_norm": 2.57100248336792, | |
| "learning_rate": 5.45e-05, | |
| "loss": 0.7383, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 58.48329048843188, | |
| "eval_accuracy": 0.8045938243522368, | |
| "eval_loss": 0.9126896858215332, | |
| "eval_runtime": 33.0001, | |
| "eval_samples_per_second": 188.575, | |
| "eval_steps_per_second": 5.909, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 59.12596401028278, | |
| "grad_norm": 2.6267364025115967, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.7372, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 59.12596401028278, | |
| "eval_accuracy": 0.8047429982319032, | |
| "eval_loss": 0.9037800431251526, | |
| "eval_runtime": 33.4956, | |
| "eval_samples_per_second": 185.786, | |
| "eval_steps_per_second": 5.822, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 59.768637532133674, | |
| "grad_norm": 2.6067042350769043, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 0.7293, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 59.768637532133674, | |
| "eval_accuracy": 0.8063738822187712, | |
| "eval_loss": 0.891742467880249, | |
| "eval_runtime": 33.3465, | |
| "eval_samples_per_second": 186.616, | |
| "eval_steps_per_second": 5.848, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 60.411311053984576, | |
| "grad_norm": 2.748093843460083, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 0.7266, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 60.411311053984576, | |
| "eval_accuracy": 0.8060851226391891, | |
| "eval_loss": 0.8950145840644836, | |
| "eval_runtime": 33.417, | |
| "eval_samples_per_second": 186.223, | |
| "eval_steps_per_second": 5.835, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 61.05398457583548, | |
| "grad_norm": 2.7752370834350586, | |
| "learning_rate": 5.25e-05, | |
| "loss": 0.7278, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 61.05398457583548, | |
| "eval_accuracy": 0.8052570807938081, | |
| "eval_loss": 0.9001266956329346, | |
| "eval_runtime": 33.2575, | |
| "eval_samples_per_second": 187.116, | |
| "eval_steps_per_second": 5.863, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 61.696658097686374, | |
| "grad_norm": 2.4993159770965576, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.7224, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 61.696658097686374, | |
| "eval_accuracy": 0.8063545774342619, | |
| "eval_loss": 0.9000985622406006, | |
| "eval_runtime": 32.5061, | |
| "eval_samples_per_second": 191.441, | |
| "eval_steps_per_second": 5.999, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 62.339331619537276, | |
| "grad_norm": 2.795344829559326, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 0.7156, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 62.339331619537276, | |
| "eval_accuracy": 0.8078125860121053, | |
| "eval_loss": 0.8850185871124268, | |
| "eval_runtime": 33.3065, | |
| "eval_samples_per_second": 186.84, | |
| "eval_steps_per_second": 5.855, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 62.98200514138817, | |
| "grad_norm": 2.4736196994781494, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 0.7179, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 62.98200514138817, | |
| "eval_accuracy": 0.8069766099061457, | |
| "eval_loss": 0.8914857506752014, | |
| "eval_runtime": 33.2699, | |
| "eval_samples_per_second": 187.046, | |
| "eval_steps_per_second": 5.861, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 63.624678663239074, | |
| "grad_norm": 2.7739675045013428, | |
| "learning_rate": 5.05e-05, | |
| "loss": 0.7158, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 63.624678663239074, | |
| "eval_accuracy": 0.806712037699331, | |
| "eval_loss": 0.8928523063659668, | |
| "eval_runtime": 33.5382, | |
| "eval_samples_per_second": 185.55, | |
| "eval_steps_per_second": 5.814, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 64.26735218508998, | |
| "grad_norm": 2.6932947635650635, | |
| "learning_rate": 5e-05, | |
| "loss": 0.711, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 64.26735218508998, | |
| "eval_accuracy": 0.8060159735552406, | |
| "eval_loss": 0.9013872742652893, | |
| "eval_runtime": 33.9301, | |
| "eval_samples_per_second": 183.407, | |
| "eval_steps_per_second": 5.747, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 64.91002570694087, | |
| "grad_norm": 2.7735486030578613, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.7097, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 64.91002570694087, | |
| "eval_accuracy": 0.8067475980060699, | |
| "eval_loss": 0.8957981467247009, | |
| "eval_runtime": 32.2026, | |
| "eval_samples_per_second": 193.245, | |
| "eval_steps_per_second": 6.055, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 65.55269922879178, | |
| "grad_norm": 2.9156525135040283, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.7079, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 65.55269922879178, | |
| "eval_accuracy": 0.8081258363288117, | |
| "eval_loss": 0.894716739654541, | |
| "eval_runtime": 32.7636, | |
| "eval_samples_per_second": 189.936, | |
| "eval_steps_per_second": 5.952, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 66.19537275064268, | |
| "grad_norm": 2.6359803676605225, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.7091, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 66.19537275064268, | |
| "eval_accuracy": 0.8079719372684523, | |
| "eval_loss": 0.8968275189399719, | |
| "eval_runtime": 32.9482, | |
| "eval_samples_per_second": 188.872, | |
| "eval_steps_per_second": 5.918, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 66.83804627249357, | |
| "grad_norm": 2.481441020965576, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.7023, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 66.83804627249357, | |
| "eval_accuracy": 0.8089100382369007, | |
| "eval_loss": 0.8855522274971008, | |
| "eval_runtime": 33.2751, | |
| "eval_samples_per_second": 187.017, | |
| "eval_steps_per_second": 5.86, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 67.48071979434447, | |
| "grad_norm": 2.7268731594085693, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.701, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 67.48071979434447, | |
| "eval_accuracy": 0.8084944735097863, | |
| "eval_loss": 0.8864369988441467, | |
| "eval_runtime": 33.1906, | |
| "eval_samples_per_second": 187.493, | |
| "eval_steps_per_second": 5.875, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 68.12339331619538, | |
| "grad_norm": 2.5931150913238525, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6988, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 68.12339331619538, | |
| "eval_accuracy": 0.8089927462608983, | |
| "eval_loss": 0.8866479992866516, | |
| "eval_runtime": 33.2058, | |
| "eval_samples_per_second": 187.407, | |
| "eval_steps_per_second": 5.872, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 68.76606683804627, | |
| "grad_norm": 2.736328601837158, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.6994, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 68.76606683804627, | |
| "eval_accuracy": 0.8084927024019897, | |
| "eval_loss": 0.8820142149925232, | |
| "eval_runtime": 32.9508, | |
| "eval_samples_per_second": 188.857, | |
| "eval_steps_per_second": 5.918, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 69.40874035989717, | |
| "grad_norm": 2.8003087043762207, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6831, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 69.40874035989717, | |
| "eval_accuracy": 0.8091836520937638, | |
| "eval_loss": 0.8856648206710815, | |
| "eval_runtime": 33.3733, | |
| "eval_samples_per_second": 186.466, | |
| "eval_steps_per_second": 5.843, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 70.05141388174808, | |
| "grad_norm": 2.7809290885925293, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.6948, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 70.05141388174808, | |
| "eval_accuracy": 0.8089959519265054, | |
| "eval_loss": 0.89335036277771, | |
| "eval_runtime": 33.324, | |
| "eval_samples_per_second": 186.743, | |
| "eval_steps_per_second": 5.852, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 70.69408740359897, | |
| "grad_norm": 2.5319223403930664, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.688, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 70.69408740359897, | |
| "eval_accuracy": 0.8102897851773414, | |
| "eval_loss": 0.8859269022941589, | |
| "eval_runtime": 33.0875, | |
| "eval_samples_per_second": 188.077, | |
| "eval_steps_per_second": 5.893, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 71.33676092544987, | |
| "grad_norm": 3.0281503200531006, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.6894, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 71.33676092544987, | |
| "eval_accuracy": 0.8097594530479655, | |
| "eval_loss": 0.8875166773796082, | |
| "eval_runtime": 32.7248, | |
| "eval_samples_per_second": 190.161, | |
| "eval_steps_per_second": 5.959, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 71.97943444730078, | |
| "grad_norm": 2.476452589035034, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6855, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 71.97943444730078, | |
| "eval_accuracy": 0.8097316597458161, | |
| "eval_loss": 0.8884576559066772, | |
| "eval_runtime": 33.1154, | |
| "eval_samples_per_second": 187.919, | |
| "eval_steps_per_second": 5.889, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 72.62210796915167, | |
| "grad_norm": 2.662523031234741, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.6811, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 72.62210796915167, | |
| "eval_accuracy": 0.809297297864581, | |
| "eval_loss": 0.8833766579627991, | |
| "eval_runtime": 33.1424, | |
| "eval_samples_per_second": 187.766, | |
| "eval_steps_per_second": 5.884, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 73.26478149100257, | |
| "grad_norm": 2.4902071952819824, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6782, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 73.26478149100257, | |
| "eval_accuracy": 0.8089903844893186, | |
| "eval_loss": 0.8910095691680908, | |
| "eval_runtime": 33.3324, | |
| "eval_samples_per_second": 186.695, | |
| "eval_steps_per_second": 5.85, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 73.90745501285348, | |
| "grad_norm": 2.897806167602539, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.6761, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 73.90745501285348, | |
| "eval_accuracy": 0.8115941649425137, | |
| "eval_loss": 0.8758607506752014, | |
| "eval_runtime": 33.305, | |
| "eval_samples_per_second": 186.849, | |
| "eval_steps_per_second": 5.855, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 74.55012853470437, | |
| "grad_norm": 2.6833012104034424, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.6713, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 74.55012853470437, | |
| "eval_accuracy": 0.8105924384594244, | |
| "eval_loss": 0.8842668533325195, | |
| "eval_runtime": 32.104, | |
| "eval_samples_per_second": 193.839, | |
| "eval_steps_per_second": 6.074, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 75.19280205655527, | |
| "grad_norm": 3.058183193206787, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.6751, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 75.19280205655527, | |
| "eval_accuracy": 0.8100959331379869, | |
| "eval_loss": 0.8922275304794312, | |
| "eval_runtime": 32.0158, | |
| "eval_samples_per_second": 194.373, | |
| "eval_steps_per_second": 6.091, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 75.83547557840618, | |
| "grad_norm": 2.5572476387023926, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.667, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 75.83547557840618, | |
| "eval_accuracy": 0.8115087833220307, | |
| "eval_loss": 0.8827089071273804, | |
| "eval_runtime": 32.068, | |
| "eval_samples_per_second": 194.056, | |
| "eval_steps_per_second": 6.081, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 76.47814910025707, | |
| "grad_norm": 2.6060643196105957, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.6628, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 76.47814910025707, | |
| "eval_accuracy": 0.8114595604343708, | |
| "eval_loss": 0.8741211891174316, | |
| "eval_runtime": 32.0679, | |
| "eval_samples_per_second": 194.057, | |
| "eval_steps_per_second": 6.081, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 77.12082262210797, | |
| "grad_norm": 2.643742322921753, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6685, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 77.12082262210797, | |
| "eval_accuracy": 0.8120777359560294, | |
| "eval_loss": 0.8767244219779968, | |
| "eval_runtime": 32.012, | |
| "eval_samples_per_second": 194.396, | |
| "eval_steps_per_second": 6.091, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 77.76349614395887, | |
| "grad_norm": 3.0320351123809814, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.6618, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 77.76349614395887, | |
| "eval_accuracy": 0.8108977044235925, | |
| "eval_loss": 0.882882833480835, | |
| "eval_runtime": 32.0023, | |
| "eval_samples_per_second": 194.455, | |
| "eval_steps_per_second": 6.093, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 78.40616966580977, | |
| "grad_norm": 2.645836114883423, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.6626, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 78.40616966580977, | |
| "eval_accuracy": 0.8128617042506485, | |
| "eval_loss": 0.8741999268531799, | |
| "eval_runtime": 32.0443, | |
| "eval_samples_per_second": 194.2, | |
| "eval_steps_per_second": 6.085, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 79.04884318766067, | |
| "grad_norm": 2.858215570449829, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.6617, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 79.04884318766067, | |
| "eval_accuracy": 0.8136581954693509, | |
| "eval_loss": 0.8712067604064941, | |
| "eval_runtime": 32.061, | |
| "eval_samples_per_second": 194.099, | |
| "eval_steps_per_second": 6.082, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 79.69151670951157, | |
| "grad_norm": 2.5988807678222656, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.655, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 79.69151670951157, | |
| "eval_accuracy": 0.8128823572620334, | |
| "eval_loss": 0.8742080926895142, | |
| "eval_runtime": 32.0565, | |
| "eval_samples_per_second": 194.126, | |
| "eval_steps_per_second": 6.083, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 80.33419023136247, | |
| "grad_norm": 2.6401941776275635, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.6545, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 80.33419023136247, | |
| "eval_accuracy": 0.812843042566724, | |
| "eval_loss": 0.8718605041503906, | |
| "eval_runtime": 31.9943, | |
| "eval_samples_per_second": 194.503, | |
| "eval_steps_per_second": 6.095, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 80.97686375321337, | |
| "grad_norm": 2.8062636852264404, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.655, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 80.97686375321337, | |
| "eval_accuracy": 0.811865863240631, | |
| "eval_loss": 0.8775948882102966, | |
| "eval_runtime": 32.0581, | |
| "eval_samples_per_second": 194.116, | |
| "eval_steps_per_second": 6.083, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 81.61953727506426, | |
| "grad_norm": 2.548220157623291, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.6468, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 81.61953727506426, | |
| "eval_accuracy": 0.8119920345795418, | |
| "eval_loss": 0.883579432964325, | |
| "eval_runtime": 32.0753, | |
| "eval_samples_per_second": 194.012, | |
| "eval_steps_per_second": 6.079, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 82.26221079691517, | |
| "grad_norm": 2.5861518383026123, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6487, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 82.26221079691517, | |
| "eval_accuracy": 0.8131239363417762, | |
| "eval_loss": 0.8730902075767517, | |
| "eval_runtime": 32.1045, | |
| "eval_samples_per_second": 193.835, | |
| "eval_steps_per_second": 6.074, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 82.90488431876607, | |
| "grad_norm": 2.780217409133911, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.6495, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 82.90488431876607, | |
| "eval_accuracy": 0.8124547391400174, | |
| "eval_loss": 0.8819155097007751, | |
| "eval_runtime": 32.0183, | |
| "eval_samples_per_second": 194.357, | |
| "eval_steps_per_second": 6.09, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 83.54755784061696, | |
| "grad_norm": 2.5522782802581787, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.6425, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 83.54755784061696, | |
| "eval_accuracy": 0.8124381288595807, | |
| "eval_loss": 0.885295569896698, | |
| "eval_runtime": 32.0293, | |
| "eval_samples_per_second": 194.291, | |
| "eval_steps_per_second": 6.088, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 84.19023136246787, | |
| "grad_norm": 2.6785833835601807, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.6423, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 84.19023136246787, | |
| "eval_accuracy": 0.8132641437128263, | |
| "eval_loss": 0.8791692852973938, | |
| "eval_runtime": 32.0427, | |
| "eval_samples_per_second": 194.209, | |
| "eval_steps_per_second": 6.086, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 84.83290488431876, | |
| "grad_norm": 2.6489408016204834, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.6377, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 84.83290488431876, | |
| "eval_accuracy": 0.8154822401853656, | |
| "eval_loss": 0.8657127618789673, | |
| "eval_runtime": 32.102, | |
| "eval_samples_per_second": 193.851, | |
| "eval_steps_per_second": 6.074, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 85.47557840616966, | |
| "grad_norm": 2.556199312210083, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.6334, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 85.47557840616966, | |
| "eval_accuracy": 0.8141076970581924, | |
| "eval_loss": 0.8747490644454956, | |
| "eval_runtime": 32.0636, | |
| "eval_samples_per_second": 194.083, | |
| "eval_steps_per_second": 6.082, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 86.11825192802057, | |
| "grad_norm": 2.477529764175415, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.6421, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 86.11825192802057, | |
| "eval_accuracy": 0.8136067662347515, | |
| "eval_loss": 0.8816725015640259, | |
| "eval_runtime": 32.0966, | |
| "eval_samples_per_second": 193.883, | |
| "eval_steps_per_second": 6.075, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 86.76092544987146, | |
| "grad_norm": 2.6774518489837646, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.6366, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 86.76092544987146, | |
| "eval_accuracy": 0.814412672148727, | |
| "eval_loss": 0.8695808053016663, | |
| "eval_runtime": 32.0613, | |
| "eval_samples_per_second": 194.097, | |
| "eval_steps_per_second": 6.082, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 87.40359897172236, | |
| "grad_norm": 2.6455330848693848, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.6305, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 87.40359897172236, | |
| "eval_accuracy": 0.8148675345064433, | |
| "eval_loss": 0.8716031312942505, | |
| "eval_runtime": 32.0969, | |
| "eval_samples_per_second": 193.882, | |
| "eval_steps_per_second": 6.075, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 88.04627249357327, | |
| "grad_norm": 2.7725086212158203, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.6325, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 88.04627249357327, | |
| "eval_accuracy": 0.813131776644422, | |
| "eval_loss": 0.8786645531654358, | |
| "eval_runtime": 32.024, | |
| "eval_samples_per_second": 194.323, | |
| "eval_steps_per_second": 6.089, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 88.68894601542416, | |
| "grad_norm": 2.6008856296539307, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.6282, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 88.68894601542416, | |
| "eval_accuracy": 0.8146181108967369, | |
| "eval_loss": 0.8678516149520874, | |
| "eval_runtime": 32.0037, | |
| "eval_samples_per_second": 194.447, | |
| "eval_steps_per_second": 6.093, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 89.33161953727506, | |
| "grad_norm": 2.7128331661224365, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.6261, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 89.33161953727506, | |
| "eval_accuracy": 0.8144260367762641, | |
| "eval_loss": 0.871059000492096, | |
| "eval_runtime": 32.0173, | |
| "eval_samples_per_second": 194.364, | |
| "eval_steps_per_second": 6.09, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 89.97429305912597, | |
| "grad_norm": 2.324878692626953, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6263, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 89.97429305912597, | |
| "eval_accuracy": 0.8156161510225485, | |
| "eval_loss": 0.8654680252075195, | |
| "eval_runtime": 32.0295, | |
| "eval_samples_per_second": 194.29, | |
| "eval_steps_per_second": 6.088, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 90.61696658097686, | |
| "grad_norm": 2.7399654388427734, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.6233, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 90.61696658097686, | |
| "eval_accuracy": 0.8162172670337523, | |
| "eval_loss": 0.8661888241767883, | |
| "eval_runtime": 32.0203, | |
| "eval_samples_per_second": 194.346, | |
| "eval_steps_per_second": 6.09, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 91.25964010282776, | |
| "grad_norm": 2.8856687545776367, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.6176, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 91.25964010282776, | |
| "eval_accuracy": 0.8161184520753214, | |
| "eval_loss": 0.8683423399925232, | |
| "eval_runtime": 32.0579, | |
| "eval_samples_per_second": 194.117, | |
| "eval_steps_per_second": 6.083, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 91.90231362467867, | |
| "grad_norm": 2.4138548374176025, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.6239, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 91.90231362467867, | |
| "eval_accuracy": 0.8153887598602306, | |
| "eval_loss": 0.8599680066108704, | |
| "eval_runtime": 32.0297, | |
| "eval_samples_per_second": 194.289, | |
| "eval_steps_per_second": 6.088, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 92.54498714652956, | |
| "grad_norm": 2.6343512535095215, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.6189, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 92.54498714652956, | |
| "eval_accuracy": 0.8154344802246543, | |
| "eval_loss": 0.8673732876777649, | |
| "eval_runtime": 32.0317, | |
| "eval_samples_per_second": 194.276, | |
| "eval_steps_per_second": 6.088, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 93.18766066838046, | |
| "grad_norm": 2.6886157989501953, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.6117, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 93.18766066838046, | |
| "eval_accuracy": 0.8162594874166894, | |
| "eval_loss": 0.8670706748962402, | |
| "eval_runtime": 32.0796, | |
| "eval_samples_per_second": 193.986, | |
| "eval_steps_per_second": 6.079, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 93.83033419023137, | |
| "grad_norm": 2.6037755012512207, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.615, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 93.83033419023137, | |
| "eval_accuracy": 0.8150874471307802, | |
| "eval_loss": 0.8663885593414307, | |
| "eval_runtime": 32.0301, | |
| "eval_samples_per_second": 194.286, | |
| "eval_steps_per_second": 6.088, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 94.47300771208226, | |
| "grad_norm": 2.5215036869049072, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.6138, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 94.47300771208226, | |
| "eval_accuracy": 0.8168731185953577, | |
| "eval_loss": 0.8630892634391785, | |
| "eval_runtime": 32.066, | |
| "eval_samples_per_second": 194.069, | |
| "eval_steps_per_second": 6.081, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 95.11568123393316, | |
| "grad_norm": 2.797325611114502, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.6107, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 95.11568123393316, | |
| "eval_accuracy": 0.8156241335474724, | |
| "eval_loss": 0.8718482851982117, | |
| "eval_runtime": 32.0559, | |
| "eval_samples_per_second": 194.13, | |
| "eval_steps_per_second": 6.083, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 95.75835475578407, | |
| "grad_norm": 3.0883982181549072, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.6099, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 95.75835475578407, | |
| "eval_accuracy": 0.8160560285491784, | |
| "eval_loss": 0.8676818013191223, | |
| "eval_runtime": 32.0504, | |
| "eval_samples_per_second": 194.163, | |
| "eval_steps_per_second": 6.084, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 96.40102827763496, | |
| "grad_norm": 2.922588348388672, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6026, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 96.40102827763496, | |
| "eval_accuracy": 0.8162867029734353, | |
| "eval_loss": 0.8721068501472473, | |
| "eval_runtime": 32.019, | |
| "eval_samples_per_second": 194.353, | |
| "eval_steps_per_second": 6.09, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 97.04370179948586, | |
| "grad_norm": 2.663724660873413, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.6084, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 97.04370179948586, | |
| "eval_accuracy": 0.8182687960636296, | |
| "eval_loss": 0.858277440071106, | |
| "eval_runtime": 32.0344, | |
| "eval_samples_per_second": 194.26, | |
| "eval_steps_per_second": 6.087, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 97.68637532133675, | |
| "grad_norm": 2.8198440074920654, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.6033, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 97.68637532133675, | |
| "eval_accuracy": 0.8163839511643823, | |
| "eval_loss": 0.8688974380493164, | |
| "eval_runtime": 32.0629, | |
| "eval_samples_per_second": 194.087, | |
| "eval_steps_per_second": 6.082, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 98.32904884318766, | |
| "grad_norm": 2.3979711532592773, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.603, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 98.32904884318766, | |
| "eval_accuracy": 0.8169446192573946, | |
| "eval_loss": 0.8678939342498779, | |
| "eval_runtime": 32.0422, | |
| "eval_samples_per_second": 194.213, | |
| "eval_steps_per_second": 6.086, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 98.97172236503856, | |
| "grad_norm": 2.643324613571167, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.6009, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 98.97172236503856, | |
| "eval_accuracy": 0.8176424081803114, | |
| "eval_loss": 0.8609100580215454, | |
| "eval_runtime": 32.0681, | |
| "eval_samples_per_second": 194.056, | |
| "eval_steps_per_second": 6.081, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 99.61439588688945, | |
| "grad_norm": 2.791355848312378, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.601, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 99.61439588688945, | |
| "eval_accuracy": 0.816467387038539, | |
| "eval_loss": 0.867489755153656, | |
| "eval_runtime": 32.1192, | |
| "eval_samples_per_second": 193.747, | |
| "eval_steps_per_second": 6.071, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 100.25706940874036, | |
| "grad_norm": 2.787290096282959, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.5966, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 100.25706940874036, | |
| "eval_accuracy": 0.8171603328039044, | |
| "eval_loss": 0.8652510046958923, | |
| "eval_runtime": 32.0539, | |
| "eval_samples_per_second": 194.142, | |
| "eval_steps_per_second": 6.083, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 100.89974293059126, | |
| "grad_norm": 2.791646957397461, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.5935, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 100.89974293059126, | |
| "eval_accuracy": 0.8178409455191369, | |
| "eval_loss": 0.8600234985351562, | |
| "eval_runtime": 32.0149, | |
| "eval_samples_per_second": 194.378, | |
| "eval_steps_per_second": 6.091, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 101.54241645244215, | |
| "grad_norm": 2.4990010261535645, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.5912, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 101.54241645244215, | |
| "eval_accuracy": 0.8169466975666281, | |
| "eval_loss": 0.8653005957603455, | |
| "eval_runtime": 31.9972, | |
| "eval_samples_per_second": 194.486, | |
| "eval_steps_per_second": 6.094, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 102.18508997429306, | |
| "grad_norm": 2.3002467155456543, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.5873, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 102.18508997429306, | |
| "eval_accuracy": 0.81857837810817, | |
| "eval_loss": 0.8576123118400574, | |
| "eval_runtime": 31.9904, | |
| "eval_samples_per_second": 194.527, | |
| "eval_steps_per_second": 6.096, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 102.82776349614396, | |
| "grad_norm": 2.712336540222168, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5913, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 102.82776349614396, | |
| "eval_accuracy": 0.8185828450767534, | |
| "eval_loss": 0.855664074420929, | |
| "eval_runtime": 32.0488, | |
| "eval_samples_per_second": 194.173, | |
| "eval_steps_per_second": 6.084, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 103.47043701799485, | |
| "grad_norm": 2.769562005996704, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.5908, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 103.47043701799485, | |
| "eval_accuracy": 0.8184711461830915, | |
| "eval_loss": 0.8605988621711731, | |
| "eval_runtime": 32.0286, | |
| "eval_samples_per_second": 194.295, | |
| "eval_steps_per_second": 6.088, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 104.11311053984576, | |
| "grad_norm": 2.6929941177368164, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.5898, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 104.11311053984576, | |
| "eval_accuracy": 0.8183153166169729, | |
| "eval_loss": 0.8647379875183105, | |
| "eval_runtime": 32.0673, | |
| "eval_samples_per_second": 194.06, | |
| "eval_steps_per_second": 6.081, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 104.75578406169666, | |
| "grad_norm": 2.8258962631225586, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.5841, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 104.75578406169666, | |
| "eval_accuracy": 0.817774798787225, | |
| "eval_loss": 0.8651922941207886, | |
| "eval_runtime": 32.011, | |
| "eval_samples_per_second": 194.402, | |
| "eval_steps_per_second": 6.092, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 105.39845758354755, | |
| "grad_norm": 2.46081805229187, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.582, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 105.39845758354755, | |
| "eval_accuracy": 0.817711216701606, | |
| "eval_loss": 0.8619120717048645, | |
| "eval_runtime": 32.0215, | |
| "eval_samples_per_second": 194.338, | |
| "eval_steps_per_second": 6.09, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 106.04113110539846, | |
| "grad_norm": 2.960406541824341, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.5833, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 106.04113110539846, | |
| "eval_accuracy": 0.8198030559915176, | |
| "eval_loss": 0.8566803932189941, | |
| "eval_runtime": 31.9884, | |
| "eval_samples_per_second": 194.539, | |
| "eval_steps_per_second": 6.096, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 106.68380462724936, | |
| "grad_norm": 2.3195433616638184, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.5804, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 106.68380462724936, | |
| "eval_accuracy": 0.8193473285144606, | |
| "eval_loss": 0.8546783328056335, | |
| "eval_runtime": 32.0522, | |
| "eval_samples_per_second": 194.152, | |
| "eval_steps_per_second": 6.084, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 107.32647814910025, | |
| "grad_norm": 2.7828962802886963, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.5839, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 107.32647814910025, | |
| "eval_accuracy": 0.817869892847695, | |
| "eval_loss": 0.859340488910675, | |
| "eval_runtime": 32.2415, | |
| "eval_samples_per_second": 193.012, | |
| "eval_steps_per_second": 6.048, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 107.96915167095116, | |
| "grad_norm": 2.8258416652679443, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.5774, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 107.96915167095116, | |
| "eval_accuracy": 0.8181858662056681, | |
| "eval_loss": 0.8593936562538147, | |
| "eval_runtime": 32.0552, | |
| "eval_samples_per_second": 194.134, | |
| "eval_steps_per_second": 6.083, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 108.61182519280206, | |
| "grad_norm": 2.5473225116729736, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.5791, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 108.61182519280206, | |
| "eval_accuracy": 0.8201209689038143, | |
| "eval_loss": 0.8578335046768188, | |
| "eval_runtime": 32.0789, | |
| "eval_samples_per_second": 193.99, | |
| "eval_steps_per_second": 6.079, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 109.25449871465295, | |
| "grad_norm": 2.6014022827148438, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.5724, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 109.25449871465295, | |
| "eval_accuracy": 0.8179463406457561, | |
| "eval_loss": 0.8615684509277344, | |
| "eval_runtime": 32.0619, | |
| "eval_samples_per_second": 194.093, | |
| "eval_steps_per_second": 6.082, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 109.89717223650386, | |
| "grad_norm": 2.6044790744781494, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.5782, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 109.89717223650386, | |
| "eval_accuracy": 0.818407341792689, | |
| "eval_loss": 0.8595022559165955, | |
| "eval_runtime": 32.0145, | |
| "eval_samples_per_second": 194.381, | |
| "eval_steps_per_second": 6.091, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 110.53984575835476, | |
| "grad_norm": 2.570466995239258, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.5735, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 110.53984575835476, | |
| "eval_accuracy": 0.8203102481461101, | |
| "eval_loss": 0.8511986136436462, | |
| "eval_runtime": 32.0544, | |
| "eval_samples_per_second": 194.139, | |
| "eval_steps_per_second": 6.083, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 111.18251928020565, | |
| "grad_norm": 2.4701082706451416, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.5664, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 111.18251928020565, | |
| "eval_accuracy": 0.8206346877390821, | |
| "eval_loss": 0.8540939688682556, | |
| "eval_runtime": 32.046, | |
| "eval_samples_per_second": 194.19, | |
| "eval_steps_per_second": 6.085, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 111.82519280205655, | |
| "grad_norm": 2.9233691692352295, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.5719, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 111.82519280205655, | |
| "eval_accuracy": 0.8200771187996747, | |
| "eval_loss": 0.8584414124488831, | |
| "eval_runtime": 32.0359, | |
| "eval_samples_per_second": 194.251, | |
| "eval_steps_per_second": 6.087, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 112.46786632390746, | |
| "grad_norm": 2.3986382484436035, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.5682, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 112.46786632390746, | |
| "eval_accuracy": 0.8193907912255869, | |
| "eval_loss": 0.8565849661827087, | |
| "eval_runtime": 32.0525, | |
| "eval_samples_per_second": 194.15, | |
| "eval_steps_per_second": 6.084, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 113.11053984575835, | |
| "grad_norm": 2.5226054191589355, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.5719, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 113.11053984575835, | |
| "eval_accuracy": 0.8199272051177008, | |
| "eval_loss": 0.8500083684921265, | |
| "eval_runtime": 32.0889, | |
| "eval_samples_per_second": 193.93, | |
| "eval_steps_per_second": 6.077, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 113.75321336760925, | |
| "grad_norm": 2.609858274459839, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.5674, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 113.75321336760925, | |
| "eval_accuracy": 0.8205847363486674, | |
| "eval_loss": 0.8524363040924072, | |
| "eval_runtime": 32.0335, | |
| "eval_samples_per_second": 194.265, | |
| "eval_steps_per_second": 6.087, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 114.39588688946016, | |
| "grad_norm": 2.2834887504577637, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.5643, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 114.39588688946016, | |
| "eval_accuracy": 0.821479995279122, | |
| "eval_loss": 0.8454416990280151, | |
| "eval_runtime": 32.0112, | |
| "eval_samples_per_second": 194.401, | |
| "eval_steps_per_second": 6.092, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 115.03856041131105, | |
| "grad_norm": 2.776373863220215, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.5681, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 115.03856041131105, | |
| "eval_accuracy": 0.8196546091505466, | |
| "eval_loss": 0.8588145971298218, | |
| "eval_runtime": 31.2312, | |
| "eval_samples_per_second": 199.256, | |
| "eval_steps_per_second": 6.244, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 115.68123393316195, | |
| "grad_norm": 2.7792084217071533, | |
| "learning_rate": 1e-05, | |
| "loss": 0.563, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 115.68123393316195, | |
| "eval_accuracy": 0.8215367161011942, | |
| "eval_loss": 0.8472273945808411, | |
| "eval_runtime": 32.0865, | |
| "eval_samples_per_second": 193.944, | |
| "eval_steps_per_second": 6.077, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 116.32390745501286, | |
| "grad_norm": 2.5021698474884033, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.5563, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 116.32390745501286, | |
| "eval_accuracy": 0.8198419059293134, | |
| "eval_loss": 0.8576837778091431, | |
| "eval_runtime": 32.0421, | |
| "eval_samples_per_second": 194.214, | |
| "eval_steps_per_second": 6.086, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 116.96658097686375, | |
| "grad_norm": 2.7690136432647705, | |
| "learning_rate": 9e-06, | |
| "loss": 0.5667, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 116.96658097686375, | |
| "eval_accuracy": 0.8201571685631533, | |
| "eval_loss": 0.8592654466629028, | |
| "eval_runtime": 32.0402, | |
| "eval_samples_per_second": 194.225, | |
| "eval_steps_per_second": 6.086, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 117.60925449871465, | |
| "grad_norm": 2.7082719802856445, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.5598, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 117.60925449871465, | |
| "eval_accuracy": 0.8194785991809854, | |
| "eval_loss": 0.8624646067619324, | |
| "eval_runtime": 31.9751, | |
| "eval_samples_per_second": 194.62, | |
| "eval_steps_per_second": 6.098, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 118.25192802056556, | |
| "grad_norm": 2.6081254482269287, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.5588, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 118.25192802056556, | |
| "eval_accuracy": 0.8200990472992823, | |
| "eval_loss": 0.8558657169342041, | |
| "eval_runtime": 32.0709, | |
| "eval_samples_per_second": 194.039, | |
| "eval_steps_per_second": 6.08, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 118.89460154241645, | |
| "grad_norm": 2.39003849029541, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.5573, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 118.89460154241645, | |
| "eval_accuracy": 0.821544695178437, | |
| "eval_loss": 0.8529016375541687, | |
| "eval_runtime": 32.0688, | |
| "eval_samples_per_second": 194.052, | |
| "eval_steps_per_second": 6.081, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 119.53727506426735, | |
| "grad_norm": 2.671393871307373, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.5583, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 119.53727506426735, | |
| "eval_accuracy": 0.8202954907050746, | |
| "eval_loss": 0.8513576984405518, | |
| "eval_runtime": 32.0535, | |
| "eval_samples_per_second": 194.144, | |
| "eval_steps_per_second": 6.084, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 120.17994858611826, | |
| "grad_norm": 2.8136119842529297, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.5599, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 120.17994858611826, | |
| "eval_accuracy": 0.8210522728285633, | |
| "eval_loss": 0.8533715605735779, | |
| "eval_runtime": 32.0108, | |
| "eval_samples_per_second": 194.403, | |
| "eval_steps_per_second": 6.092, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 120.82262210796915, | |
| "grad_norm": 2.5622072219848633, | |
| "learning_rate": 6e-06, | |
| "loss": 0.5538, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 120.82262210796915, | |
| "eval_accuracy": 0.8210836727544613, | |
| "eval_loss": 0.8567976951599121, | |
| "eval_runtime": 32.0891, | |
| "eval_samples_per_second": 193.929, | |
| "eval_steps_per_second": 6.077, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 121.46529562982005, | |
| "grad_norm": 2.870851516723633, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.5544, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 121.46529562982005, | |
| "eval_accuracy": 0.8227272131920523, | |
| "eval_loss": 0.8419870138168335, | |
| "eval_runtime": 32.0445, | |
| "eval_samples_per_second": 194.199, | |
| "eval_steps_per_second": 6.085, | |
| "step": 94500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 129, | |
| "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": 7.979699566574305e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |