| { | |
| "best_metric": 0.7950243949890137, | |
| "best_model_checkpoint": "./model_fine-tune/glot/xlm-r/kat-Geor/checkpoint-99500", | |
| "epoch": 32.34720416124838, | |
| "eval_steps": 500, | |
| "global_step": 99500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1625487646293888, | |
| "grad_norm": 4.931751251220703, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.3006, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1625487646293888, | |
| "eval_accuracy": 0.7623556146386671, | |
| "eval_loss": 1.3419054746627808, | |
| "eval_runtime": 224.5592, | |
| "eval_samples_per_second": 90.996, | |
| "eval_steps_per_second": 2.846, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3250975292587776, | |
| "grad_norm": 4.145026683807373, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.2154, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3250975292587776, | |
| "eval_accuracy": 0.7721005611086652, | |
| "eval_loss": 1.2983847856521606, | |
| "eval_runtime": 225.9654, | |
| "eval_samples_per_second": 90.43, | |
| "eval_steps_per_second": 2.828, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48764629388816644, | |
| "grad_norm": 4.589369297027588, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.1779, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.48764629388816644, | |
| "eval_accuracy": 0.7790585887983502, | |
| "eval_loss": 1.2703238725662231, | |
| "eval_runtime": 226.7078, | |
| "eval_samples_per_second": 90.134, | |
| "eval_steps_per_second": 2.819, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6501950585175552, | |
| "grad_norm": 4.552799701690674, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.1546, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6501950585175552, | |
| "eval_accuracy": 0.7819611016961427, | |
| "eval_loss": 1.2609333992004395, | |
| "eval_runtime": 226.665, | |
| "eval_samples_per_second": 90.151, | |
| "eval_steps_per_second": 2.819, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.812743823146944, | |
| "grad_norm": 3.709968328475952, | |
| "learning_rate": 9.75e-05, | |
| "loss": 1.1259, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.812743823146944, | |
| "eval_accuracy": 0.7877356734428504, | |
| "eval_loss": 1.2131903171539307, | |
| "eval_runtime": 226.8657, | |
| "eval_samples_per_second": 90.071, | |
| "eval_steps_per_second": 2.817, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.9752925877763329, | |
| "grad_norm": 3.6233718395233154, | |
| "learning_rate": 9.7e-05, | |
| "loss": 1.0938, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9752925877763329, | |
| "eval_accuracy": 0.789409001949858, | |
| "eval_loss": 1.2194353342056274, | |
| "eval_runtime": 226.5843, | |
| "eval_samples_per_second": 90.183, | |
| "eval_steps_per_second": 2.82, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.1378413524057218, | |
| "grad_norm": 3.7706315517425537, | |
| "learning_rate": 9.65e-05, | |
| "loss": 1.0728, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.1378413524057218, | |
| "eval_accuracy": 0.7930909730559011, | |
| "eval_loss": 1.2004605531692505, | |
| "eval_runtime": 225.3386, | |
| "eval_samples_per_second": 90.681, | |
| "eval_steps_per_second": 2.836, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.3003901170351106, | |
| "grad_norm": 3.654762029647827, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.067, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.3003901170351106, | |
| "eval_accuracy": 0.7911667822625772, | |
| "eval_loss": 1.206479549407959, | |
| "eval_runtime": 229.9442, | |
| "eval_samples_per_second": 88.865, | |
| "eval_steps_per_second": 2.779, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.4629388816644995, | |
| "grad_norm": 3.6047844886779785, | |
| "learning_rate": 9.55e-05, | |
| "loss": 1.0623, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.4629388816644995, | |
| "eval_accuracy": 0.7950484617863337, | |
| "eval_loss": 1.186224341392517, | |
| "eval_runtime": 229.5346, | |
| "eval_samples_per_second": 89.024, | |
| "eval_steps_per_second": 2.784, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.6254876462938883, | |
| "grad_norm": 3.174414873123169, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.0452, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.6254876462938883, | |
| "eval_accuracy": 0.7986192006290228, | |
| "eval_loss": 1.1624418497085571, | |
| "eval_runtime": 227.8114, | |
| "eval_samples_per_second": 89.697, | |
| "eval_steps_per_second": 2.805, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.7880364109232771, | |
| "grad_norm": 3.455960988998413, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 1.0203, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.7880364109232771, | |
| "eval_accuracy": 0.7996825782405662, | |
| "eval_loss": 1.160246729850769, | |
| "eval_runtime": 227.8469, | |
| "eval_samples_per_second": 89.683, | |
| "eval_steps_per_second": 2.805, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.9505851755526658, | |
| "grad_norm": 3.4478368759155273, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.0102, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.9505851755526658, | |
| "eval_accuracy": 0.8028851426121806, | |
| "eval_loss": 1.1382070779800415, | |
| "eval_runtime": 225.8399, | |
| "eval_samples_per_second": 90.48, | |
| "eval_steps_per_second": 2.829, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.113133940182055, | |
| "grad_norm": 3.5076828002929688, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 1.0007, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.113133940182055, | |
| "eval_accuracy": 0.8036015201468353, | |
| "eval_loss": 1.1308486461639404, | |
| "eval_runtime": 227.3542, | |
| "eval_samples_per_second": 89.877, | |
| "eval_steps_per_second": 2.811, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.2756827048114436, | |
| "grad_norm": 2.954378366470337, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 0.9828, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.2756827048114436, | |
| "eval_accuracy": 0.80609597689472, | |
| "eval_loss": 1.1300016641616821, | |
| "eval_runtime": 226.3258, | |
| "eval_samples_per_second": 90.286, | |
| "eval_steps_per_second": 2.823, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.438231469440832, | |
| "grad_norm": 3.6220898628234863, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 0.9738, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.438231469440832, | |
| "eval_accuracy": 0.807549206838832, | |
| "eval_loss": 1.1173433065414429, | |
| "eval_runtime": 230.0766, | |
| "eval_samples_per_second": 88.814, | |
| "eval_steps_per_second": 2.777, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.6007802340702213, | |
| "grad_norm": 3.0354886054992676, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.9726, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.6007802340702213, | |
| "eval_accuracy": 0.8092077277721563, | |
| "eval_loss": 1.1192331314086914, | |
| "eval_runtime": 228.747, | |
| "eval_samples_per_second": 89.33, | |
| "eval_steps_per_second": 2.793, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.7633289986996097, | |
| "grad_norm": 2.905066967010498, | |
| "learning_rate": 9.15e-05, | |
| "loss": 0.9639, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.7633289986996097, | |
| "eval_accuracy": 0.8105359765383991, | |
| "eval_loss": 1.0964252948760986, | |
| "eval_runtime": 239.0977, | |
| "eval_samples_per_second": 85.463, | |
| "eval_steps_per_second": 2.673, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.925877763328999, | |
| "grad_norm": 2.5866756439208984, | |
| "learning_rate": 9.1e-05, | |
| "loss": 0.9562, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.925877763328999, | |
| "eval_accuracy": 0.8124049786000702, | |
| "eval_loss": 1.0979543924331665, | |
| "eval_runtime": 228.0298, | |
| "eval_samples_per_second": 89.611, | |
| "eval_steps_per_second": 2.802, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.0884265279583873, | |
| "grad_norm": 3.0912091732025146, | |
| "learning_rate": 9.05e-05, | |
| "loss": 0.9412, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.0884265279583873, | |
| "eval_accuracy": 0.8128664613154557, | |
| "eval_loss": 1.098233699798584, | |
| "eval_runtime": 226.3594, | |
| "eval_samples_per_second": 90.272, | |
| "eval_steps_per_second": 2.823, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.250975292587776, | |
| "grad_norm": 3.310755968093872, | |
| "learning_rate": 9e-05, | |
| "loss": 0.9404, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.250975292587776, | |
| "eval_accuracy": 0.8141734154155298, | |
| "eval_loss": 1.0834821462631226, | |
| "eval_runtime": 229.2889, | |
| "eval_samples_per_second": 89.119, | |
| "eval_steps_per_second": 2.787, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.413524057217165, | |
| "grad_norm": 3.2035183906555176, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 0.9345, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.413524057217165, | |
| "eval_accuracy": 0.8148430925920552, | |
| "eval_loss": 1.071053385734558, | |
| "eval_runtime": 228.6307, | |
| "eval_samples_per_second": 89.376, | |
| "eval_steps_per_second": 2.795, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.576072821846554, | |
| "grad_norm": 3.3350119590759277, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 0.9287, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.576072821846554, | |
| "eval_accuracy": 0.8148937367897325, | |
| "eval_loss": 1.0869466066360474, | |
| "eval_runtime": 229.0653, | |
| "eval_samples_per_second": 89.206, | |
| "eval_steps_per_second": 2.79, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.7386215864759427, | |
| "grad_norm": 3.0126075744628906, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 0.916, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.7386215864759427, | |
| "eval_accuracy": 0.8164226040931482, | |
| "eval_loss": 1.0795150995254517, | |
| "eval_runtime": 227.3062, | |
| "eval_samples_per_second": 89.896, | |
| "eval_steps_per_second": 2.811, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.9011703511053315, | |
| "grad_norm": 2.9025819301605225, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.9114, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.9011703511053315, | |
| "eval_accuracy": 0.8172531893844662, | |
| "eval_loss": 1.0739566087722778, | |
| "eval_runtime": 227.5695, | |
| "eval_samples_per_second": 89.792, | |
| "eval_steps_per_second": 2.808, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.06371911573472, | |
| "grad_norm": 3.1317036151885986, | |
| "learning_rate": 8.75e-05, | |
| "loss": 0.9104, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.06371911573472, | |
| "eval_accuracy": 0.8181712807844302, | |
| "eval_loss": 1.057742953300476, | |
| "eval_runtime": 226.8271, | |
| "eval_samples_per_second": 90.086, | |
| "eval_steps_per_second": 2.817, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.22626788036411, | |
| "grad_norm": 2.9879837036132812, | |
| "learning_rate": 8.7e-05, | |
| "loss": 0.8932, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.22626788036411, | |
| "eval_accuracy": 0.819528608544276, | |
| "eval_loss": 1.0524249076843262, | |
| "eval_runtime": 227.5673, | |
| "eval_samples_per_second": 89.793, | |
| "eval_steps_per_second": 2.808, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 4.388816644993498, | |
| "grad_norm": 3.1617486476898193, | |
| "learning_rate": 8.65e-05, | |
| "loss": 0.8884, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.388816644993498, | |
| "eval_accuracy": 0.8202382159332025, | |
| "eval_loss": 1.0618681907653809, | |
| "eval_runtime": 228.7279, | |
| "eval_samples_per_second": 89.338, | |
| "eval_steps_per_second": 2.794, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.551365409622887, | |
| "grad_norm": 3.287351131439209, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.8932, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.551365409622887, | |
| "eval_accuracy": 0.821493227612126, | |
| "eval_loss": 1.0429145097732544, | |
| "eval_runtime": 229.2177, | |
| "eval_samples_per_second": 89.147, | |
| "eval_steps_per_second": 2.788, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.713914174252276, | |
| "grad_norm": 3.1148979663848877, | |
| "learning_rate": 8.55e-05, | |
| "loss": 0.8871, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.713914174252276, | |
| "eval_accuracy": 0.821055764186474, | |
| "eval_loss": 1.0499247312545776, | |
| "eval_runtime": 227.9461, | |
| "eval_samples_per_second": 89.644, | |
| "eval_steps_per_second": 2.803, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.876462938881664, | |
| "grad_norm": 3.0099728107452393, | |
| "learning_rate": 8.5e-05, | |
| "loss": 0.8848, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.876462938881664, | |
| "eval_accuracy": 0.8221091492282949, | |
| "eval_loss": 1.0302263498306274, | |
| "eval_runtime": 229.0246, | |
| "eval_samples_per_second": 89.222, | |
| "eval_steps_per_second": 2.79, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.039011703511053, | |
| "grad_norm": 2.8249874114990234, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 0.8772, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 5.039011703511053, | |
| "eval_accuracy": 0.8225020975532324, | |
| "eval_loss": 1.0407395362854004, | |
| "eval_runtime": 230.4007, | |
| "eval_samples_per_second": 88.689, | |
| "eval_steps_per_second": 2.773, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 5.201560468140442, | |
| "grad_norm": 2.273552179336548, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.8715, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.201560468140442, | |
| "eval_accuracy": 0.8238052823384368, | |
| "eval_loss": 1.0325181484222412, | |
| "eval_runtime": 230.5347, | |
| "eval_samples_per_second": 88.637, | |
| "eval_steps_per_second": 2.772, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 5.364109232769831, | |
| "grad_norm": 2.884051561355591, | |
| "learning_rate": 8.35e-05, | |
| "loss": 0.8687, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.364109232769831, | |
| "eval_accuracy": 0.8243404153748926, | |
| "eval_loss": 1.0298871994018555, | |
| "eval_runtime": 231.4091, | |
| "eval_samples_per_second": 88.302, | |
| "eval_steps_per_second": 2.761, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.526657997399219, | |
| "grad_norm": 3.12709379196167, | |
| "learning_rate": 8.3e-05, | |
| "loss": 0.8674, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.526657997399219, | |
| "eval_accuracy": 0.8245251637171149, | |
| "eval_loss": 1.0233651399612427, | |
| "eval_runtime": 225.4766, | |
| "eval_samples_per_second": 90.626, | |
| "eval_steps_per_second": 2.834, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.689206762028609, | |
| "grad_norm": 2.7715749740600586, | |
| "learning_rate": 8.25e-05, | |
| "loss": 0.8597, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.689206762028609, | |
| "eval_accuracy": 0.8253348922363006, | |
| "eval_loss": 1.0265984535217285, | |
| "eval_runtime": 226.4158, | |
| "eval_samples_per_second": 90.25, | |
| "eval_steps_per_second": 2.822, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.851755526657997, | |
| "grad_norm": 2.630910873413086, | |
| "learning_rate": 8.2e-05, | |
| "loss": 0.8621, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.851755526657997, | |
| "eval_accuracy": 0.8264286938861036, | |
| "eval_loss": 1.025406002998352, | |
| "eval_runtime": 225.265, | |
| "eval_samples_per_second": 90.711, | |
| "eval_steps_per_second": 2.837, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.014304291287386, | |
| "grad_norm": 2.960498571395874, | |
| "learning_rate": 8.15e-05, | |
| "loss": 0.8505, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 6.014304291287386, | |
| "eval_accuracy": 0.8266802737227764, | |
| "eval_loss": 1.0281902551651, | |
| "eval_runtime": 225.5813, | |
| "eval_samples_per_second": 90.584, | |
| "eval_steps_per_second": 2.833, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 6.176853055916775, | |
| "grad_norm": 2.7250943183898926, | |
| "learning_rate": 8.1e-05, | |
| "loss": 0.8434, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.176853055916775, | |
| "eval_accuracy": 0.8273285977403005, | |
| "eval_loss": 1.019544005393982, | |
| "eval_runtime": 225.1972, | |
| "eval_samples_per_second": 90.738, | |
| "eval_steps_per_second": 2.838, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 6.339401820546164, | |
| "grad_norm": 2.45147967338562, | |
| "learning_rate": 8.05e-05, | |
| "loss": 0.8396, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.339401820546164, | |
| "eval_accuracy": 0.8278276606616609, | |
| "eval_loss": 1.016022801399231, | |
| "eval_runtime": 225.1071, | |
| "eval_samples_per_second": 90.775, | |
| "eval_steps_per_second": 2.839, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 6.501950585175552, | |
| "grad_norm": 3.0252089500427246, | |
| "learning_rate": 8e-05, | |
| "loss": 0.849, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.501950585175552, | |
| "eval_accuracy": 0.8286753807254127, | |
| "eval_loss": 0.9984493851661682, | |
| "eval_runtime": 225.6293, | |
| "eval_samples_per_second": 90.564, | |
| "eval_steps_per_second": 2.832, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.664499349804942, | |
| "grad_norm": 2.79874849319458, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 0.8344, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.664499349804942, | |
| "eval_accuracy": 0.8295993429023929, | |
| "eval_loss": 1.008083462715149, | |
| "eval_runtime": 226.4394, | |
| "eval_samples_per_second": 90.24, | |
| "eval_steps_per_second": 2.822, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.82704811443433, | |
| "grad_norm": 2.726980447769165, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 0.832, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.82704811443433, | |
| "eval_accuracy": 0.8289772346442087, | |
| "eval_loss": 1.0009534358978271, | |
| "eval_runtime": 225.2727, | |
| "eval_samples_per_second": 90.708, | |
| "eval_steps_per_second": 2.837, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.989596879063719, | |
| "grad_norm": 2.6177260875701904, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 0.8371, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.989596879063719, | |
| "eval_accuracy": 0.828972862081629, | |
| "eval_loss": 1.007731556892395, | |
| "eval_runtime": 225.1985, | |
| "eval_samples_per_second": 90.738, | |
| "eval_steps_per_second": 2.837, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 7.152145643693108, | |
| "grad_norm": 2.511810541152954, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.8221, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.152145643693108, | |
| "eval_accuracy": 0.8296968824079836, | |
| "eval_loss": 1.001420259475708, | |
| "eval_runtime": 226.3788, | |
| "eval_samples_per_second": 90.265, | |
| "eval_steps_per_second": 2.823, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 7.314694408322497, | |
| "grad_norm": 2.490196466445923, | |
| "learning_rate": 7.75e-05, | |
| "loss": 0.819, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.314694408322497, | |
| "eval_accuracy": 0.8315370661967273, | |
| "eval_loss": 1.0057367086410522, | |
| "eval_runtime": 226.9835, | |
| "eval_samples_per_second": 90.024, | |
| "eval_steps_per_second": 2.815, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 7.477243172951885, | |
| "grad_norm": 2.7740466594696045, | |
| "learning_rate": 7.7e-05, | |
| "loss": 0.8211, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.477243172951885, | |
| "eval_accuracy": 0.8325044106908798, | |
| "eval_loss": 0.9906172752380371, | |
| "eval_runtime": 226.1902, | |
| "eval_samples_per_second": 90.34, | |
| "eval_steps_per_second": 2.825, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 7.639791937581275, | |
| "grad_norm": 3.0789926052093506, | |
| "learning_rate": 7.65e-05, | |
| "loss": 0.818, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.639791937581275, | |
| "eval_accuracy": 0.832777941957673, | |
| "eval_loss": 0.9794951677322388, | |
| "eval_runtime": 226.5187, | |
| "eval_samples_per_second": 90.209, | |
| "eval_steps_per_second": 2.821, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.802340702210663, | |
| "grad_norm": 2.452514171600342, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.8128, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.802340702210663, | |
| "eval_accuracy": 0.8325327815535476, | |
| "eval_loss": 0.9824967384338379, | |
| "eval_runtime": 225.9154, | |
| "eval_samples_per_second": 90.45, | |
| "eval_steps_per_second": 2.828, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.964889466840052, | |
| "grad_norm": 2.486287832260132, | |
| "learning_rate": 7.55e-05, | |
| "loss": 0.8133, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.964889466840052, | |
| "eval_accuracy": 0.8329719850198115, | |
| "eval_loss": 0.9840763211250305, | |
| "eval_runtime": 225.7676, | |
| "eval_samples_per_second": 90.509, | |
| "eval_steps_per_second": 2.83, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 8.12743823146944, | |
| "grad_norm": 2.6238162517547607, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.7972, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.12743823146944, | |
| "eval_accuracy": 0.8329068936503207, | |
| "eval_loss": 0.9786662459373474, | |
| "eval_runtime": 227.0649, | |
| "eval_samples_per_second": 89.992, | |
| "eval_steps_per_second": 2.814, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 8.289986996098829, | |
| "grad_norm": 2.622385025024414, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 0.8012, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 8.289986996098829, | |
| "eval_accuracy": 0.83353766380846, | |
| "eval_loss": 0.9908860325813293, | |
| "eval_runtime": 225.4511, | |
| "eval_samples_per_second": 90.636, | |
| "eval_steps_per_second": 2.834, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 8.45253576072822, | |
| "grad_norm": 2.3659889698028564, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.8002, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 8.45253576072822, | |
| "eval_accuracy": 0.834269520268432, | |
| "eval_loss": 0.9886682629585266, | |
| "eval_runtime": 226.2835, | |
| "eval_samples_per_second": 90.303, | |
| "eval_steps_per_second": 2.824, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 8.615084525357608, | |
| "grad_norm": 2.6340997219085693, | |
| "learning_rate": 7.35e-05, | |
| "loss": 0.796, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.615084525357608, | |
| "eval_accuracy": 0.83431812146113, | |
| "eval_loss": 1.1027251482009888, | |
| "eval_runtime": 224.7718, | |
| "eval_samples_per_second": 90.91, | |
| "eval_steps_per_second": 2.843, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 8.777633289986996, | |
| "grad_norm": 2.351562976837158, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.8032, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.777633289986996, | |
| "eval_accuracy": 0.83473517326524, | |
| "eval_loss": 0.9701533317565918, | |
| "eval_runtime": 224.7228, | |
| "eval_samples_per_second": 90.93, | |
| "eval_steps_per_second": 2.844, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.940182054616384, | |
| "grad_norm": 3.179107189178467, | |
| "learning_rate": 7.25e-05, | |
| "loss": 0.7956, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.940182054616384, | |
| "eval_accuracy": 0.8351845716261748, | |
| "eval_loss": 0.978928804397583, | |
| "eval_runtime": 226.3807, | |
| "eval_samples_per_second": 90.264, | |
| "eval_steps_per_second": 2.823, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 9.102730819245775, | |
| "grad_norm": 2.7947020530700684, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.7897, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 9.102730819245775, | |
| "eval_accuracy": 0.8353527515006416, | |
| "eval_loss": 0.9703938961029053, | |
| "eval_runtime": 227.4785, | |
| "eval_samples_per_second": 89.828, | |
| "eval_steps_per_second": 2.809, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 9.265279583875163, | |
| "grad_norm": 2.4941184520721436, | |
| "learning_rate": 7.15e-05, | |
| "loss": 0.7822, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 9.265279583875163, | |
| "eval_accuracy": 0.8354081248952335, | |
| "eval_loss": 0.985792875289917, | |
| "eval_runtime": 225.177, | |
| "eval_samples_per_second": 90.746, | |
| "eval_steps_per_second": 2.838, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 9.427828348504551, | |
| "grad_norm": 2.623500108718872, | |
| "learning_rate": 7.1e-05, | |
| "loss": 0.7885, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 9.427828348504551, | |
| "eval_accuracy": 0.8367873538892325, | |
| "eval_loss": 0.9710731506347656, | |
| "eval_runtime": 224.2999, | |
| "eval_samples_per_second": 91.101, | |
| "eval_steps_per_second": 2.849, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 9.59037711313394, | |
| "grad_norm": 2.687804698944092, | |
| "learning_rate": 7.05e-05, | |
| "loss": 0.7853, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 9.59037711313394, | |
| "eval_accuracy": 0.8370918540482961, | |
| "eval_loss": 0.9488577246665955, | |
| "eval_runtime": 225.5849, | |
| "eval_samples_per_second": 90.582, | |
| "eval_steps_per_second": 2.833, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 9.752925877763328, | |
| "grad_norm": 2.415875196456909, | |
| "learning_rate": 7e-05, | |
| "loss": 0.7834, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.752925877763328, | |
| "eval_accuracy": 0.8376393865857097, | |
| "eval_loss": 0.9489056468009949, | |
| "eval_runtime": 224.5743, | |
| "eval_samples_per_second": 90.99, | |
| "eval_steps_per_second": 2.845, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.915474642392718, | |
| "grad_norm": 2.635620355606079, | |
| "learning_rate": 6.95e-05, | |
| "loss": 0.7795, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 9.915474642392718, | |
| "eval_accuracy": 0.8379052346488646, | |
| "eval_loss": 0.9600683450698853, | |
| "eval_runtime": 227.5569, | |
| "eval_samples_per_second": 89.797, | |
| "eval_steps_per_second": 2.808, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 10.078023407022107, | |
| "grad_norm": 2.2729740142822266, | |
| "learning_rate": 6.9e-05, | |
| "loss": 0.7784, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 10.078023407022107, | |
| "eval_accuracy": 0.8387711597381718, | |
| "eval_loss": 0.9676991701126099, | |
| "eval_runtime": 227.0441, | |
| "eval_samples_per_second": 90.0, | |
| "eval_steps_per_second": 2.814, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 10.240572171651495, | |
| "grad_norm": 2.665194511413574, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 0.7731, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 10.240572171651495, | |
| "eval_accuracy": 0.8384039380286287, | |
| "eval_loss": 0.9660561084747314, | |
| "eval_runtime": 225.7122, | |
| "eval_samples_per_second": 90.531, | |
| "eval_steps_per_second": 2.831, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 10.403120936280883, | |
| "grad_norm": 2.378326892852783, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.7732, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 10.403120936280883, | |
| "eval_accuracy": 0.8384703620413777, | |
| "eval_loss": 0.9522095322608948, | |
| "eval_runtime": 226.2237, | |
| "eval_samples_per_second": 90.327, | |
| "eval_steps_per_second": 2.825, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 10.565669700910274, | |
| "grad_norm": 2.3873043060302734, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 0.7675, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 10.565669700910274, | |
| "eval_accuracy": 0.8387303829039866, | |
| "eval_loss": 0.9506338238716125, | |
| "eval_runtime": 225.3705, | |
| "eval_samples_per_second": 90.668, | |
| "eval_steps_per_second": 2.835, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 10.728218465539662, | |
| "grad_norm": 2.908627986907959, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.7632, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 10.728218465539662, | |
| "eval_accuracy": 0.8393342612931991, | |
| "eval_loss": 0.9406708478927612, | |
| "eval_runtime": 226.5163, | |
| "eval_samples_per_second": 90.21, | |
| "eval_steps_per_second": 2.821, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 10.89076723016905, | |
| "grad_norm": 2.353163242340088, | |
| "learning_rate": 6.65e-05, | |
| "loss": 0.7652, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 10.89076723016905, | |
| "eval_accuracy": 0.8397924279467366, | |
| "eval_loss": 0.9450008869171143, | |
| "eval_runtime": 228.8699, | |
| "eval_samples_per_second": 89.282, | |
| "eval_steps_per_second": 2.792, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 11.053315994798439, | |
| "grad_norm": 2.5943007469177246, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.7569, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 11.053315994798439, | |
| "eval_accuracy": 0.8403287583785509, | |
| "eval_loss": 0.9399783611297607, | |
| "eval_runtime": 226.1121, | |
| "eval_samples_per_second": 90.371, | |
| "eval_steps_per_second": 2.826, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 11.215864759427829, | |
| "grad_norm": 2.693864107131958, | |
| "learning_rate": 6.55e-05, | |
| "loss": 0.7594, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 11.215864759427829, | |
| "eval_accuracy": 0.8401560936743129, | |
| "eval_loss": 0.9480236172676086, | |
| "eval_runtime": 225.4517, | |
| "eval_samples_per_second": 90.636, | |
| "eval_steps_per_second": 2.834, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 11.378413524057217, | |
| "grad_norm": 2.644212484359741, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.7549, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 11.378413524057217, | |
| "eval_accuracy": 0.8397807491674708, | |
| "eval_loss": 0.947754442691803, | |
| "eval_runtime": 224.7666, | |
| "eval_samples_per_second": 90.912, | |
| "eval_steps_per_second": 2.843, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 11.540962288686606, | |
| "grad_norm": 2.548055410385132, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 0.7508, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 11.540962288686606, | |
| "eval_accuracy": 0.8408923535971057, | |
| "eval_loss": 0.9363004565238953, | |
| "eval_runtime": 226.2948, | |
| "eval_samples_per_second": 90.298, | |
| "eval_steps_per_second": 2.824, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 11.703511053315994, | |
| "grad_norm": 2.2983460426330566, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.7498, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 11.703511053315994, | |
| "eval_accuracy": 0.8415923695179955, | |
| "eval_loss": 0.9559971690177917, | |
| "eval_runtime": 225.5228, | |
| "eval_samples_per_second": 90.607, | |
| "eval_steps_per_second": 2.833, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 11.866059817945384, | |
| "grad_norm": 2.585132122039795, | |
| "learning_rate": 6.35e-05, | |
| "loss": 0.7529, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 11.866059817945384, | |
| "eval_accuracy": 0.8414500365516611, | |
| "eval_loss": 0.9397149085998535, | |
| "eval_runtime": 225.5184, | |
| "eval_samples_per_second": 90.609, | |
| "eval_steps_per_second": 2.833, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 12.028608582574773, | |
| "grad_norm": 2.7008893489837646, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.7499, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 12.028608582574773, | |
| "eval_accuracy": 0.8429048722887895, | |
| "eval_loss": 0.9322752952575684, | |
| "eval_runtime": 226.0814, | |
| "eval_samples_per_second": 90.383, | |
| "eval_steps_per_second": 2.826, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 12.191157347204161, | |
| "grad_norm": 2.234344959259033, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.7516, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 12.191157347204161, | |
| "eval_accuracy": 0.8423553938504648, | |
| "eval_loss": 0.9274143576622009, | |
| "eval_runtime": 224.6544, | |
| "eval_samples_per_second": 90.957, | |
| "eval_steps_per_second": 2.844, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 12.35370611183355, | |
| "grad_norm": 2.38130259513855, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.742, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 12.35370611183355, | |
| "eval_accuracy": 0.8434239606449652, | |
| "eval_loss": 0.9298020005226135, | |
| "eval_runtime": 224.8881, | |
| "eval_samples_per_second": 90.863, | |
| "eval_steps_per_second": 2.841, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 12.51625487646294, | |
| "grad_norm": 2.4753472805023193, | |
| "learning_rate": 6.15e-05, | |
| "loss": 0.7349, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 12.51625487646294, | |
| "eval_accuracy": 0.8435835943839733, | |
| "eval_loss": 0.9265833497047424, | |
| "eval_runtime": 225.4856, | |
| "eval_samples_per_second": 90.622, | |
| "eval_steps_per_second": 2.834, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 12.678803641092328, | |
| "grad_norm": 2.6576738357543945, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.7408, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 12.678803641092328, | |
| "eval_accuracy": 0.8433502485604528, | |
| "eval_loss": 0.9311428070068359, | |
| "eval_runtime": 226.2881, | |
| "eval_samples_per_second": 90.301, | |
| "eval_steps_per_second": 2.824, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 12.841352405721716, | |
| "grad_norm": 2.318998336791992, | |
| "learning_rate": 6.05e-05, | |
| "loss": 0.742, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 12.841352405721716, | |
| "eval_accuracy": 0.8439347116351804, | |
| "eval_loss": 0.9278241991996765, | |
| "eval_runtime": 226.17, | |
| "eval_samples_per_second": 90.348, | |
| "eval_steps_per_second": 2.825, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 13.003901170351105, | |
| "grad_norm": 2.5888776779174805, | |
| "learning_rate": 6e-05, | |
| "loss": 0.735, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 13.003901170351105, | |
| "eval_accuracy": 0.8435309371055906, | |
| "eval_loss": 0.9364249110221863, | |
| "eval_runtime": 226.5769, | |
| "eval_samples_per_second": 90.186, | |
| "eval_steps_per_second": 2.82, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 13.166449934980495, | |
| "grad_norm": 2.5382964611053467, | |
| "learning_rate": 5.95e-05, | |
| "loss": 0.7317, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 13.166449934980495, | |
| "eval_accuracy": 0.8435685143232774, | |
| "eval_loss": 0.9289753437042236, | |
| "eval_runtime": 226.0444, | |
| "eval_samples_per_second": 90.398, | |
| "eval_steps_per_second": 2.827, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 13.328998699609883, | |
| "grad_norm": 2.3561949729919434, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.7287, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 13.328998699609883, | |
| "eval_accuracy": 0.8441468378237561, | |
| "eval_loss": 0.9176314473152161, | |
| "eval_runtime": 225.4845, | |
| "eval_samples_per_second": 90.623, | |
| "eval_steps_per_second": 2.834, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 13.491547464239272, | |
| "grad_norm": 2.4950144290924072, | |
| "learning_rate": 5.85e-05, | |
| "loss": 0.7236, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 13.491547464239272, | |
| "eval_accuracy": 0.8440490876509515, | |
| "eval_loss": 0.9126904010772705, | |
| "eval_runtime": 225.1525, | |
| "eval_samples_per_second": 90.756, | |
| "eval_steps_per_second": 2.838, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 13.65409622886866, | |
| "grad_norm": 2.5094358921051025, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.7271, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 13.65409622886866, | |
| "eval_accuracy": 0.844420508518643, | |
| "eval_loss": 0.9222070574760437, | |
| "eval_runtime": 226.8881, | |
| "eval_samples_per_second": 90.062, | |
| "eval_steps_per_second": 2.816, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 13.81664499349805, | |
| "grad_norm": 2.5358660221099854, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.7247, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 13.81664499349805, | |
| "eval_accuracy": 0.8447406385701589, | |
| "eval_loss": 0.911141037940979, | |
| "eval_runtime": 226.4876, | |
| "eval_samples_per_second": 90.221, | |
| "eval_steps_per_second": 2.821, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 13.979193758127439, | |
| "grad_norm": 2.449117422103882, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.724, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 13.979193758127439, | |
| "eval_accuracy": 0.8455180692523965, | |
| "eval_loss": 0.9233959317207336, | |
| "eval_runtime": 226.6454, | |
| "eval_samples_per_second": 90.158, | |
| "eval_steps_per_second": 2.819, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 14.141742522756827, | |
| "grad_norm": 2.9299075603485107, | |
| "learning_rate": 5.65e-05, | |
| "loss": 0.717, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 14.141742522756827, | |
| "eval_accuracy": 0.8457589583161474, | |
| "eval_loss": 0.9212149381637573, | |
| "eval_runtime": 226.3542, | |
| "eval_samples_per_second": 90.274, | |
| "eval_steps_per_second": 2.823, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 14.304291287386215, | |
| "grad_norm": 2.705927610397339, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.7209, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 14.304291287386215, | |
| "eval_accuracy": 0.8461540554265815, | |
| "eval_loss": 0.9096006155014038, | |
| "eval_runtime": 225.3071, | |
| "eval_samples_per_second": 90.694, | |
| "eval_steps_per_second": 2.836, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 14.466840052015606, | |
| "grad_norm": 2.0237460136413574, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 0.7143, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 14.466840052015606, | |
| "eval_accuracy": 0.8467122032637585, | |
| "eval_loss": 0.9058763980865479, | |
| "eval_runtime": 224.7066, | |
| "eval_samples_per_second": 90.936, | |
| "eval_steps_per_second": 2.844, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 14.629388816644994, | |
| "grad_norm": 2.6412603855133057, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.7097, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 14.629388816644994, | |
| "eval_accuracy": 0.8469896352526538, | |
| "eval_loss": 0.9109433889389038, | |
| "eval_runtime": 226.4374, | |
| "eval_samples_per_second": 90.241, | |
| "eval_steps_per_second": 2.822, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 14.791937581274382, | |
| "grad_norm": 2.0750892162323, | |
| "learning_rate": 5.45e-05, | |
| "loss": 0.7167, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 14.791937581274382, | |
| "eval_accuracy": 0.8459233510762237, | |
| "eval_loss": 0.9634975790977478, | |
| "eval_runtime": 224.623, | |
| "eval_samples_per_second": 90.97, | |
| "eval_steps_per_second": 2.845, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 14.95448634590377, | |
| "grad_norm": 3.2299835681915283, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.7175, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 14.95448634590377, | |
| "eval_accuracy": 0.8472973119877086, | |
| "eval_loss": 0.9039083123207092, | |
| "eval_runtime": 226.4661, | |
| "eval_samples_per_second": 90.23, | |
| "eval_steps_per_second": 2.822, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 15.117035110533159, | |
| "grad_norm": 2.562220573425293, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 0.7032, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 15.117035110533159, | |
| "eval_accuracy": 0.8473520786144324, | |
| "eval_loss": 0.911507785320282, | |
| "eval_runtime": 224.8892, | |
| "eval_samples_per_second": 90.863, | |
| "eval_steps_per_second": 2.841, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 15.27958387516255, | |
| "grad_norm": 2.8594844341278076, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 0.7082, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 15.27958387516255, | |
| "eval_accuracy": 0.8476350142048673, | |
| "eval_loss": 0.909827470779419, | |
| "eval_runtime": 225.7006, | |
| "eval_samples_per_second": 90.536, | |
| "eval_steps_per_second": 2.831, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 15.442132639791938, | |
| "grad_norm": 2.468425989151001, | |
| "learning_rate": 5.25e-05, | |
| "loss": 0.7041, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 15.442132639791938, | |
| "eval_accuracy": 0.8478791054352419, | |
| "eval_loss": 0.9049926996231079, | |
| "eval_runtime": 224.8771, | |
| "eval_samples_per_second": 90.867, | |
| "eval_steps_per_second": 2.842, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 15.604681404421326, | |
| "grad_norm": 2.389726400375366, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.7031, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 15.604681404421326, | |
| "eval_accuracy": 0.8481061886894397, | |
| "eval_loss": 0.890941858291626, | |
| "eval_runtime": 226.989, | |
| "eval_samples_per_second": 90.022, | |
| "eval_steps_per_second": 2.815, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 15.767230169050714, | |
| "grad_norm": 2.2124507427215576, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 0.6986, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 15.767230169050714, | |
| "eval_accuracy": 0.8491361761433549, | |
| "eval_loss": 0.9069699048995972, | |
| "eval_runtime": 224.9486, | |
| "eval_samples_per_second": 90.839, | |
| "eval_steps_per_second": 2.841, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 15.929778933680105, | |
| "grad_norm": 2.2906134128570557, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 0.707, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 15.929778933680105, | |
| "eval_accuracy": 0.8479968642310016, | |
| "eval_loss": 0.9036478996276855, | |
| "eval_runtime": 225.9102, | |
| "eval_samples_per_second": 90.452, | |
| "eval_steps_per_second": 2.829, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 16.092327698309493, | |
| "grad_norm": 2.6677193641662598, | |
| "learning_rate": 5.05e-05, | |
| "loss": 0.6948, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 16.092327698309493, | |
| "eval_accuracy": 0.8493272111938716, | |
| "eval_loss": 0.9056188464164734, | |
| "eval_runtime": 225.0003, | |
| "eval_samples_per_second": 90.818, | |
| "eval_steps_per_second": 2.84, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 16.25487646293888, | |
| "grad_norm": 2.4262144565582275, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6967, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 16.25487646293888, | |
| "eval_accuracy": 0.8485872336104957, | |
| "eval_loss": 0.896799623966217, | |
| "eval_runtime": 224.1595, | |
| "eval_samples_per_second": 91.158, | |
| "eval_steps_per_second": 2.851, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 16.41742522756827, | |
| "grad_norm": 2.316122055053711, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.6977, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 16.41742522756827, | |
| "eval_accuracy": 0.8495556837644054, | |
| "eval_loss": 0.9065157175064087, | |
| "eval_runtime": 225.8926, | |
| "eval_samples_per_second": 90.459, | |
| "eval_steps_per_second": 2.829, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 16.579973992197658, | |
| "grad_norm": 2.6691653728485107, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6961, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 16.579973992197658, | |
| "eval_accuracy": 0.8500652800696321, | |
| "eval_loss": 0.896096408367157, | |
| "eval_runtime": 232.2659, | |
| "eval_samples_per_second": 87.977, | |
| "eval_steps_per_second": 2.751, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 16.742522756827046, | |
| "grad_norm": 2.475905656814575, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.6916, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 16.742522756827046, | |
| "eval_accuracy": 0.8497042599251818, | |
| "eval_loss": 0.8976150155067444, | |
| "eval_runtime": 226.0766, | |
| "eval_samples_per_second": 90.385, | |
| "eval_steps_per_second": 2.826, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 16.90507152145644, | |
| "grad_norm": 2.786370277404785, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6914, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 16.90507152145644, | |
| "eval_accuracy": 0.8496545695743757, | |
| "eval_loss": 0.8938505053520203, | |
| "eval_runtime": 224.6085, | |
| "eval_samples_per_second": 90.976, | |
| "eval_steps_per_second": 2.845, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 17.067620286085827, | |
| "grad_norm": 2.8185195922851562, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.6843, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 17.067620286085827, | |
| "eval_accuracy": 0.8506024205488569, | |
| "eval_loss": 0.902123749256134, | |
| "eval_runtime": 225.1853, | |
| "eval_samples_per_second": 90.743, | |
| "eval_steps_per_second": 2.838, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 17.230169050715215, | |
| "grad_norm": 2.287614107131958, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6886, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 17.230169050715215, | |
| "eval_accuracy": 0.8509516099496137, | |
| "eval_loss": 0.8869491815567017, | |
| "eval_runtime": 225.8223, | |
| "eval_samples_per_second": 90.487, | |
| "eval_steps_per_second": 2.83, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 17.392717815344604, | |
| "grad_norm": 2.545166492462158, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.6895, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 17.392717815344604, | |
| "eval_accuracy": 0.8509741315809264, | |
| "eval_loss": 0.88730388879776, | |
| "eval_runtime": 225.2327, | |
| "eval_samples_per_second": 90.724, | |
| "eval_steps_per_second": 2.837, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 17.555266579973992, | |
| "grad_norm": 2.133690357208252, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.684, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 17.555266579973992, | |
| "eval_accuracy": 0.8506065339944305, | |
| "eval_loss": 0.8796634674072266, | |
| "eval_runtime": 225.9145, | |
| "eval_samples_per_second": 90.45, | |
| "eval_steps_per_second": 2.829, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 17.71781534460338, | |
| "grad_norm": 2.368135690689087, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.6814, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 17.71781534460338, | |
| "eval_accuracy": 0.8518810256708096, | |
| "eval_loss": 0.8883038759231567, | |
| "eval_runtime": 226.0516, | |
| "eval_samples_per_second": 90.395, | |
| "eval_steps_per_second": 2.827, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 17.88036410923277, | |
| "grad_norm": 2.93131685256958, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6814, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 17.88036410923277, | |
| "eval_accuracy": 0.8515069138621577, | |
| "eval_loss": 0.8804932832717896, | |
| "eval_runtime": 224.3083, | |
| "eval_samples_per_second": 91.098, | |
| "eval_steps_per_second": 2.849, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 18.042912873862157, | |
| "grad_norm": 2.6115188598632812, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.6825, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 18.042912873862157, | |
| "eval_accuracy": 0.8524721846400168, | |
| "eval_loss": 0.8792157173156738, | |
| "eval_runtime": 225.8719, | |
| "eval_samples_per_second": 90.467, | |
| "eval_steps_per_second": 2.829, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 18.20546163849155, | |
| "grad_norm": 3.650696039199829, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6754, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 18.20546163849155, | |
| "eval_accuracy": 0.8515890870808295, | |
| "eval_loss": 0.8762602210044861, | |
| "eval_runtime": 225.4458, | |
| "eval_samples_per_second": 90.638, | |
| "eval_steps_per_second": 2.834, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 18.368010403120937, | |
| "grad_norm": 2.48811936378479, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.6694, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 18.368010403120937, | |
| "eval_accuracy": 0.852083188035357, | |
| "eval_loss": 0.8903321623802185, | |
| "eval_runtime": 231.1517, | |
| "eval_samples_per_second": 88.401, | |
| "eval_steps_per_second": 2.764, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 18.530559167750326, | |
| "grad_norm": 2.664206027984619, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.6711, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 18.530559167750326, | |
| "eval_accuracy": 0.8524635038150364, | |
| "eval_loss": 0.8883496522903442, | |
| "eval_runtime": 224.7773, | |
| "eval_samples_per_second": 90.908, | |
| "eval_steps_per_second": 2.843, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 18.693107932379714, | |
| "grad_norm": 2.5020461082458496, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.6701, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 18.693107932379714, | |
| "eval_accuracy": 0.8527774358008777, | |
| "eval_loss": 0.8737440705299377, | |
| "eval_runtime": 224.3031, | |
| "eval_samples_per_second": 91.1, | |
| "eval_steps_per_second": 2.849, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 18.855656697009103, | |
| "grad_norm": 2.341423273086548, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.6722, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 18.855656697009103, | |
| "eval_accuracy": 0.8529963243635703, | |
| "eval_loss": 0.8735175132751465, | |
| "eval_runtime": 225.4167, | |
| "eval_samples_per_second": 90.65, | |
| "eval_steps_per_second": 2.835, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 19.01820546163849, | |
| "grad_norm": 2.4827890396118164, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.6723, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 19.01820546163849, | |
| "eval_accuracy": 0.853864474597493, | |
| "eval_loss": 0.8714447021484375, | |
| "eval_runtime": 225.7934, | |
| "eval_samples_per_second": 90.499, | |
| "eval_steps_per_second": 2.83, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 19.18075422626788, | |
| "grad_norm": 2.6257293224334717, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.6623, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 19.18075422626788, | |
| "eval_accuracy": 0.8535918441752076, | |
| "eval_loss": 0.8707379698753357, | |
| "eval_runtime": 233.0012, | |
| "eval_samples_per_second": 87.699, | |
| "eval_steps_per_second": 2.742, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 19.343302990897268, | |
| "grad_norm": 2.2881245613098145, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.6629, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 19.343302990897268, | |
| "eval_accuracy": 0.8536284370228246, | |
| "eval_loss": 0.8575323820114136, | |
| "eval_runtime": 225.6251, | |
| "eval_samples_per_second": 90.566, | |
| "eval_steps_per_second": 2.832, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 19.505851755526656, | |
| "grad_norm": 2.485198497772217, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6629, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 19.505851755526656, | |
| "eval_accuracy": 0.8541063694677179, | |
| "eval_loss": 0.8660362958908081, | |
| "eval_runtime": 229.1829, | |
| "eval_samples_per_second": 89.16, | |
| "eval_steps_per_second": 2.788, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 19.668400520156048, | |
| "grad_norm": 2.365445375442505, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.6636, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 19.668400520156048, | |
| "eval_accuracy": 0.8544101305781451, | |
| "eval_loss": 0.8751095533370972, | |
| "eval_runtime": 226.4875, | |
| "eval_samples_per_second": 90.221, | |
| "eval_steps_per_second": 2.821, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 19.830949284785437, | |
| "grad_norm": 2.7341248989105225, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.6668, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 19.830949284785437, | |
| "eval_accuracy": 0.8538102982406334, | |
| "eval_loss": 0.8749545812606812, | |
| "eval_runtime": 225.3871, | |
| "eval_samples_per_second": 90.662, | |
| "eval_steps_per_second": 2.835, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 19.993498049414825, | |
| "grad_norm": 2.200584888458252, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.6658, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 19.993498049414825, | |
| "eval_accuracy": 0.8542889960446646, | |
| "eval_loss": 0.8671930432319641, | |
| "eval_runtime": 225.4174, | |
| "eval_samples_per_second": 90.65, | |
| "eval_steps_per_second": 2.835, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 20.156046814044213, | |
| "grad_norm": 2.7894246578216553, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.6616, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 20.156046814044213, | |
| "eval_accuracy": 0.8548723750905142, | |
| "eval_loss": 0.8662538528442383, | |
| "eval_runtime": 224.781, | |
| "eval_samples_per_second": 90.906, | |
| "eval_steps_per_second": 2.843, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 20.3185955786736, | |
| "grad_norm": 2.3690404891967773, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.6541, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 20.3185955786736, | |
| "eval_accuracy": 0.8550845489091575, | |
| "eval_loss": 0.865816593170166, | |
| "eval_runtime": 228.0119, | |
| "eval_samples_per_second": 89.618, | |
| "eval_steps_per_second": 2.802, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 20.48114434330299, | |
| "grad_norm": 2.6345200538635254, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.6561, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 20.48114434330299, | |
| "eval_accuracy": 0.8558111308868751, | |
| "eval_loss": 0.8506866097450256, | |
| "eval_runtime": 225.0923, | |
| "eval_samples_per_second": 90.781, | |
| "eval_steps_per_second": 2.839, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 20.64369310793238, | |
| "grad_norm": 2.1237897872924805, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.6578, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 20.64369310793238, | |
| "eval_accuracy": 0.8554877275609093, | |
| "eval_loss": 0.8685211539268494, | |
| "eval_runtime": 225.3889, | |
| "eval_samples_per_second": 90.661, | |
| "eval_steps_per_second": 2.835, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 20.806241872561767, | |
| "grad_norm": 2.37259578704834, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.657, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 20.806241872561767, | |
| "eval_accuracy": 0.8553885784782709, | |
| "eval_loss": 0.8665293455123901, | |
| "eval_runtime": 225.5872, | |
| "eval_samples_per_second": 90.581, | |
| "eval_steps_per_second": 2.833, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 20.96879063719116, | |
| "grad_norm": 2.595446825027466, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.6495, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 20.96879063719116, | |
| "eval_accuracy": 0.8558350742146078, | |
| "eval_loss": 0.8600612878799438, | |
| "eval_runtime": 225.3383, | |
| "eval_samples_per_second": 90.681, | |
| "eval_steps_per_second": 2.836, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 21.131339401820547, | |
| "grad_norm": 2.689633846282959, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.6484, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 21.131339401820547, | |
| "eval_accuracy": 0.8563225017454439, | |
| "eval_loss": 0.8686124086380005, | |
| "eval_runtime": 225.5265, | |
| "eval_samples_per_second": 90.606, | |
| "eval_steps_per_second": 2.833, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 21.293888166449936, | |
| "grad_norm": 2.4909982681274414, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.6429, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 21.293888166449936, | |
| "eval_accuracy": 0.8557131214639457, | |
| "eval_loss": 0.86305832862854, | |
| "eval_runtime": 225.2213, | |
| "eval_samples_per_second": 90.729, | |
| "eval_steps_per_second": 2.837, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 21.456436931079324, | |
| "grad_norm": 2.7555160522460938, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.6523, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 21.456436931079324, | |
| "eval_accuracy": 0.8565990504258199, | |
| "eval_loss": 0.8672294616699219, | |
| "eval_runtime": 225.3745, | |
| "eval_samples_per_second": 90.667, | |
| "eval_steps_per_second": 2.835, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 21.618985695708712, | |
| "grad_norm": 2.4315414428710938, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.6384, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 21.618985695708712, | |
| "eval_accuracy": 0.8567521705315332, | |
| "eval_loss": 0.8545786738395691, | |
| "eval_runtime": 226.5184, | |
| "eval_samples_per_second": 90.209, | |
| "eval_steps_per_second": 2.821, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 21.7815344603381, | |
| "grad_norm": 2.373359441757202, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.6512, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 21.7815344603381, | |
| "eval_accuracy": 0.8568411511443462, | |
| "eval_loss": 0.8615005612373352, | |
| "eval_runtime": 226.7483, | |
| "eval_samples_per_second": 90.118, | |
| "eval_steps_per_second": 2.818, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 21.94408322496749, | |
| "grad_norm": 2.5687010288238525, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.6452, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 21.94408322496749, | |
| "eval_accuracy": 0.856976936965993, | |
| "eval_loss": 0.8514290452003479, | |
| "eval_runtime": 224.8277, | |
| "eval_samples_per_second": 90.887, | |
| "eval_steps_per_second": 2.842, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 22.106631989596877, | |
| "grad_norm": 2.3000574111938477, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.6388, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 22.106631989596877, | |
| "eval_accuracy": 0.8576112543945463, | |
| "eval_loss": 0.8523911833763123, | |
| "eval_runtime": 225.5646, | |
| "eval_samples_per_second": 90.59, | |
| "eval_steps_per_second": 2.833, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 22.26918075422627, | |
| "grad_norm": 2.5039753913879395, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.6382, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 22.26918075422627, | |
| "eval_accuracy": 0.8567441321838508, | |
| "eval_loss": 0.8560938835144043, | |
| "eval_runtime": 226.9717, | |
| "eval_samples_per_second": 90.029, | |
| "eval_steps_per_second": 2.815, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 22.431729518855658, | |
| "grad_norm": 2.6623408794403076, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.6408, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 22.431729518855658, | |
| "eval_accuracy": 0.8576559655770156, | |
| "eval_loss": 0.854763925075531, | |
| "eval_runtime": 226.0318, | |
| "eval_samples_per_second": 90.403, | |
| "eval_steps_per_second": 2.827, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 22.594278283485046, | |
| "grad_norm": 2.3223702907562256, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.6433, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 22.594278283485046, | |
| "eval_accuracy": 0.8581642158372993, | |
| "eval_loss": 0.8585615754127502, | |
| "eval_runtime": 225.6463, | |
| "eval_samples_per_second": 90.558, | |
| "eval_steps_per_second": 2.832, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 22.756827048114435, | |
| "grad_norm": 2.545360803604126, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6371, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 22.756827048114435, | |
| "eval_accuracy": 0.8583454058165224, | |
| "eval_loss": 0.8482581973075867, | |
| "eval_runtime": 225.5921, | |
| "eval_samples_per_second": 90.579, | |
| "eval_steps_per_second": 2.833, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 22.919375812743823, | |
| "grad_norm": 2.18627667427063, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.6331, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 22.919375812743823, | |
| "eval_accuracy": 0.857949745807387, | |
| "eval_loss": 0.8507857918739319, | |
| "eval_runtime": 226.8073, | |
| "eval_samples_per_second": 90.094, | |
| "eval_steps_per_second": 2.817, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 23.08192457737321, | |
| "grad_norm": 2.8513870239257812, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.6393, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 23.08192457737321, | |
| "eval_accuracy": 0.8573411368207433, | |
| "eval_loss": 0.8502649664878845, | |
| "eval_runtime": 227.8606, | |
| "eval_samples_per_second": 89.678, | |
| "eval_steps_per_second": 2.804, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 23.2444733420026, | |
| "grad_norm": 2.961089611053467, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.6269, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 23.2444733420026, | |
| "eval_accuracy": 0.8582647966293633, | |
| "eval_loss": 0.8488523364067078, | |
| "eval_runtime": 226.4431, | |
| "eval_samples_per_second": 90.239, | |
| "eval_steps_per_second": 2.822, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 23.407022106631988, | |
| "grad_norm": 2.448005437850952, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.6284, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 23.407022106631988, | |
| "eval_accuracy": 0.8588356517259145, | |
| "eval_loss": 0.8428735136985779, | |
| "eval_runtime": 226.3587, | |
| "eval_samples_per_second": 90.273, | |
| "eval_steps_per_second": 2.823, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 23.56957087126138, | |
| "grad_norm": 2.4801974296569824, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.6311, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 23.56957087126138, | |
| "eval_accuracy": 0.8585083391102231, | |
| "eval_loss": 0.8391257524490356, | |
| "eval_runtime": 225.3955, | |
| "eval_samples_per_second": 90.658, | |
| "eval_steps_per_second": 2.835, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 23.73211963589077, | |
| "grad_norm": 2.4415576457977295, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.6333, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 23.73211963589077, | |
| "eval_accuracy": 0.8584229812472617, | |
| "eval_loss": 0.8474059104919434, | |
| "eval_runtime": 226.7984, | |
| "eval_samples_per_second": 90.098, | |
| "eval_steps_per_second": 2.817, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 23.894668400520157, | |
| "grad_norm": 2.5334153175354004, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.6291, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 23.894668400520157, | |
| "eval_accuracy": 0.8590801539747378, | |
| "eval_loss": 0.837401270866394, | |
| "eval_runtime": 225.1842, | |
| "eval_samples_per_second": 90.743, | |
| "eval_steps_per_second": 2.838, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 24.057217165149545, | |
| "grad_norm": 2.569241762161255, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.6255, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 24.057217165149545, | |
| "eval_accuracy": 0.8589290048653603, | |
| "eval_loss": 0.8332562446594238, | |
| "eval_runtime": 229.3684, | |
| "eval_samples_per_second": 89.088, | |
| "eval_steps_per_second": 2.786, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 24.219765929778934, | |
| "grad_norm": 2.526848316192627, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.6236, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 24.219765929778934, | |
| "eval_accuracy": 0.8597087528316705, | |
| "eval_loss": 0.848160445690155, | |
| "eval_runtime": 226.4635, | |
| "eval_samples_per_second": 90.231, | |
| "eval_steps_per_second": 2.822, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 24.382314694408322, | |
| "grad_norm": 2.655872106552124, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.624, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 24.382314694408322, | |
| "eval_accuracy": 0.8598462984123302, | |
| "eval_loss": 0.8461793065071106, | |
| "eval_runtime": 225.3124, | |
| "eval_samples_per_second": 90.692, | |
| "eval_steps_per_second": 2.836, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 24.54486345903771, | |
| "grad_norm": 2.109790325164795, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.6178, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 24.54486345903771, | |
| "eval_accuracy": 0.8595766101602678, | |
| "eval_loss": 0.8289109468460083, | |
| "eval_runtime": 226.5489, | |
| "eval_samples_per_second": 90.197, | |
| "eval_steps_per_second": 2.821, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 24.7074122236671, | |
| "grad_norm": 2.4060940742492676, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.6166, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 24.7074122236671, | |
| "eval_accuracy": 0.8606532868210031, | |
| "eval_loss": 0.8234532475471497, | |
| "eval_runtime": 225.3276, | |
| "eval_samples_per_second": 90.686, | |
| "eval_steps_per_second": 2.836, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 24.86996098829649, | |
| "grad_norm": 2.24013352394104, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.6247, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 24.86996098829649, | |
| "eval_accuracy": 0.8600803414966184, | |
| "eval_loss": 0.8327089548110962, | |
| "eval_runtime": 227.0917, | |
| "eval_samples_per_second": 89.981, | |
| "eval_steps_per_second": 2.814, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 25.03250975292588, | |
| "grad_norm": 2.370615005493164, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.624, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 25.03250975292588, | |
| "eval_accuracy": 0.8600759552685691, | |
| "eval_loss": 0.8379160761833191, | |
| "eval_runtime": 224.9721, | |
| "eval_samples_per_second": 90.829, | |
| "eval_steps_per_second": 2.84, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 25.195058517555267, | |
| "grad_norm": 2.192373514175415, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.6161, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 25.195058517555267, | |
| "eval_accuracy": 0.8597079760637308, | |
| "eval_loss": 0.8407602906227112, | |
| "eval_runtime": 225.6861, | |
| "eval_samples_per_second": 90.542, | |
| "eval_steps_per_second": 2.831, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 25.357607282184656, | |
| "grad_norm": 2.4751832485198975, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.6142, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 25.357607282184656, | |
| "eval_accuracy": 0.8608490074683309, | |
| "eval_loss": 0.8316988348960876, | |
| "eval_runtime": 225.4697, | |
| "eval_samples_per_second": 90.629, | |
| "eval_steps_per_second": 2.834, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 25.520156046814044, | |
| "grad_norm": 2.737602472305298, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.6193, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 25.520156046814044, | |
| "eval_accuracy": 0.860743624097801, | |
| "eval_loss": 0.8356801867485046, | |
| "eval_runtime": 226.5781, | |
| "eval_samples_per_second": 90.185, | |
| "eval_steps_per_second": 2.82, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 25.682704811443433, | |
| "grad_norm": 2.1464016437530518, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.6117, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 25.682704811443433, | |
| "eval_accuracy": 0.8607643217464414, | |
| "eval_loss": 0.8443505167961121, | |
| "eval_runtime": 225.3999, | |
| "eval_samples_per_second": 90.657, | |
| "eval_steps_per_second": 2.835, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 25.84525357607282, | |
| "grad_norm": 2.244690418243408, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.6155, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 25.84525357607282, | |
| "eval_accuracy": 0.8608062892739589, | |
| "eval_loss": 0.8309040665626526, | |
| "eval_runtime": 227.6535, | |
| "eval_samples_per_second": 89.759, | |
| "eval_steps_per_second": 2.807, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 26.00780234070221, | |
| "grad_norm": 2.2778093814849854, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6159, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 26.00780234070221, | |
| "eval_accuracy": 0.860849372733563, | |
| "eval_loss": 0.8468282222747803, | |
| "eval_runtime": 226.4415, | |
| "eval_samples_per_second": 90.24, | |
| "eval_steps_per_second": 2.822, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 26.170351105331598, | |
| "grad_norm": 2.04994535446167, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.6095, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 26.170351105331598, | |
| "eval_accuracy": 0.8617138336966451, | |
| "eval_loss": 0.8285297155380249, | |
| "eval_runtime": 224.8524, | |
| "eval_samples_per_second": 90.877, | |
| "eval_steps_per_second": 2.842, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 26.33289986996099, | |
| "grad_norm": 2.437809944152832, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.6105, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 26.33289986996099, | |
| "eval_accuracy": 0.8610518577000663, | |
| "eval_loss": 0.8272110223770142, | |
| "eval_runtime": 225.6868, | |
| "eval_samples_per_second": 90.541, | |
| "eval_steps_per_second": 2.831, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 26.495448634590378, | |
| "grad_norm": 2.2250375747680664, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.6087, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 26.495448634590378, | |
| "eval_accuracy": 0.8626399458651031, | |
| "eval_loss": 0.812351644039154, | |
| "eval_runtime": 226.8612, | |
| "eval_samples_per_second": 90.073, | |
| "eval_steps_per_second": 2.817, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 26.657997399219767, | |
| "grad_norm": 2.060518980026245, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.6151, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 26.657997399219767, | |
| "eval_accuracy": 0.8616146515518304, | |
| "eval_loss": 0.8262282609939575, | |
| "eval_runtime": 226.3417, | |
| "eval_samples_per_second": 90.279, | |
| "eval_steps_per_second": 2.823, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 26.820546163849155, | |
| "grad_norm": 2.0575308799743652, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.6095, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 26.820546163849155, | |
| "eval_accuracy": 0.8622129077230878, | |
| "eval_loss": 0.8227624893188477, | |
| "eval_runtime": 228.246, | |
| "eval_samples_per_second": 89.526, | |
| "eval_steps_per_second": 2.8, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 26.983094928478543, | |
| "grad_norm": 2.253310203552246, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.6, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 26.983094928478543, | |
| "eval_accuracy": 0.8627496487013281, | |
| "eval_loss": 0.8195393085479736, | |
| "eval_runtime": 226.7315, | |
| "eval_samples_per_second": 90.124, | |
| "eval_steps_per_second": 2.818, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 27.14564369310793, | |
| "grad_norm": 2.459129810333252, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.6013, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 27.14564369310793, | |
| "eval_accuracy": 0.8624795555933904, | |
| "eval_loss": 0.8182792663574219, | |
| "eval_runtime": 225.0241, | |
| "eval_samples_per_second": 90.808, | |
| "eval_steps_per_second": 2.84, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 27.30819245773732, | |
| "grad_norm": 2.1593105792999268, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.6001, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 27.30819245773732, | |
| "eval_accuracy": 0.8627880443457693, | |
| "eval_loss": 0.8216105103492737, | |
| "eval_runtime": 224.8609, | |
| "eval_samples_per_second": 90.874, | |
| "eval_steps_per_second": 2.842, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 27.47074122236671, | |
| "grad_norm": 2.086055040359497, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.6013, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 27.47074122236671, | |
| "eval_accuracy": 0.8621898394825607, | |
| "eval_loss": 0.8210575580596924, | |
| "eval_runtime": 226.231, | |
| "eval_samples_per_second": 90.324, | |
| "eval_steps_per_second": 2.825, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 27.6332899869961, | |
| "grad_norm": 2.5327186584472656, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.6058, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 27.6332899869961, | |
| "eval_accuracy": 0.8627229258499225, | |
| "eval_loss": 0.8185027241706848, | |
| "eval_runtime": 226.1064, | |
| "eval_samples_per_second": 90.373, | |
| "eval_steps_per_second": 2.826, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 27.79583875162549, | |
| "grad_norm": 2.3839502334594727, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.6042, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 27.79583875162549, | |
| "eval_accuracy": 0.8627582312704439, | |
| "eval_loss": 0.8231362104415894, | |
| "eval_runtime": 228.8409, | |
| "eval_samples_per_second": 89.293, | |
| "eval_steps_per_second": 2.792, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 27.958387516254877, | |
| "grad_norm": 2.195699691772461, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.5997, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 27.958387516254877, | |
| "eval_accuracy": 0.8631414086547167, | |
| "eval_loss": 0.8138222694396973, | |
| "eval_runtime": 225.1659, | |
| "eval_samples_per_second": 90.751, | |
| "eval_steps_per_second": 2.838, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 28.120936280884266, | |
| "grad_norm": 2.942133665084839, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.5976, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 28.120936280884266, | |
| "eval_accuracy": 0.8629241152120203, | |
| "eval_loss": 0.8277115225791931, | |
| "eval_runtime": 226.384, | |
| "eval_samples_per_second": 90.263, | |
| "eval_steps_per_second": 2.823, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 28.283485045513654, | |
| "grad_norm": 2.2104339599609375, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.6005, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 28.283485045513654, | |
| "eval_accuracy": 0.8635328035684335, | |
| "eval_loss": 0.8249954581260681, | |
| "eval_runtime": 225.8607, | |
| "eval_samples_per_second": 90.472, | |
| "eval_steps_per_second": 2.829, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 28.446033810143042, | |
| "grad_norm": 2.4792349338531494, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.5964, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 28.446033810143042, | |
| "eval_accuracy": 0.8631331114250733, | |
| "eval_loss": 0.8169623613357544, | |
| "eval_runtime": 224.4723, | |
| "eval_samples_per_second": 91.031, | |
| "eval_steps_per_second": 2.847, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 28.60858257477243, | |
| "grad_norm": 2.2584621906280518, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.5978, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 28.60858257477243, | |
| "eval_accuracy": 0.8637502162788443, | |
| "eval_loss": 0.8240325450897217, | |
| "eval_runtime": 226.0844, | |
| "eval_samples_per_second": 90.382, | |
| "eval_steps_per_second": 2.826, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 28.77113133940182, | |
| "grad_norm": 2.4756360054016113, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.5933, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 28.77113133940182, | |
| "eval_accuracy": 0.863286343484686, | |
| "eval_loss": 0.821010947227478, | |
| "eval_runtime": 233.6708, | |
| "eval_samples_per_second": 87.448, | |
| "eval_steps_per_second": 2.735, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 28.93368010403121, | |
| "grad_norm": 2.4466359615325928, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.595, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 28.93368010403121, | |
| "eval_accuracy": 0.8634251119288006, | |
| "eval_loss": 0.821042001247406, | |
| "eval_runtime": 225.7222, | |
| "eval_samples_per_second": 90.527, | |
| "eval_steps_per_second": 2.831, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 29.0962288686606, | |
| "grad_norm": 2.5361807346343994, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.5941, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 29.0962288686606, | |
| "eval_accuracy": 0.8634033694143568, | |
| "eval_loss": 0.8251069188117981, | |
| "eval_runtime": 226.448, | |
| "eval_samples_per_second": 90.237, | |
| "eval_steps_per_second": 2.822, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 29.258777633289988, | |
| "grad_norm": 2.5238239765167236, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5988, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 29.258777633289988, | |
| "eval_accuracy": 0.8638910896653752, | |
| "eval_loss": 0.8146407604217529, | |
| "eval_runtime": 222.0422, | |
| "eval_samples_per_second": 92.028, | |
| "eval_steps_per_second": 2.878, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 29.421326397919376, | |
| "grad_norm": 2.291041851043701, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.5923, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 29.421326397919376, | |
| "eval_accuracy": 0.8637874627536761, | |
| "eval_loss": 0.8178155422210693, | |
| "eval_runtime": 221.0933, | |
| "eval_samples_per_second": 92.423, | |
| "eval_steps_per_second": 2.89, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 29.583875162548765, | |
| "grad_norm": 2.601836681365967, | |
| "learning_rate": 9e-06, | |
| "loss": 0.5887, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 29.583875162548765, | |
| "eval_accuracy": 0.864438274609619, | |
| "eval_loss": 0.8162264823913574, | |
| "eval_runtime": 228.6953, | |
| "eval_samples_per_second": 89.35, | |
| "eval_steps_per_second": 2.794, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 29.746423927178153, | |
| "grad_norm": 2.56785249710083, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.5833, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 29.746423927178153, | |
| "eval_accuracy": 0.8640183304537222, | |
| "eval_loss": 0.8157439827919006, | |
| "eval_runtime": 224.1585, | |
| "eval_samples_per_second": 91.159, | |
| "eval_steps_per_second": 2.851, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 29.90897269180754, | |
| "grad_norm": 2.409670114517212, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.5951, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 29.90897269180754, | |
| "eval_accuracy": 0.8643441985332161, | |
| "eval_loss": 0.812107264995575, | |
| "eval_runtime": 221.533, | |
| "eval_samples_per_second": 92.239, | |
| "eval_steps_per_second": 2.884, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 30.07152145643693, | |
| "grad_norm": 2.19612717628479, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.5928, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 30.07152145643693, | |
| "eval_accuracy": 0.8640605097561019, | |
| "eval_loss": 0.8203216791152954, | |
| "eval_runtime": 221.1524, | |
| "eval_samples_per_second": 92.398, | |
| "eval_steps_per_second": 2.889, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 30.234070221066318, | |
| "grad_norm": 2.249537944793701, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.5878, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 30.234070221066318, | |
| "eval_accuracy": 0.8644494113060075, | |
| "eval_loss": 0.8103818893432617, | |
| "eval_runtime": 221.6978, | |
| "eval_samples_per_second": 92.17, | |
| "eval_steps_per_second": 2.882, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 30.39661898569571, | |
| "grad_norm": 2.63948917388916, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.5831, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 30.39661898569571, | |
| "eval_accuracy": 0.8646731645828304, | |
| "eval_loss": 0.8149096369743347, | |
| "eval_runtime": 231.6244, | |
| "eval_samples_per_second": 88.22, | |
| "eval_steps_per_second": 2.759, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 30.5591677503251, | |
| "grad_norm": 2.257843494415283, | |
| "learning_rate": 6e-06, | |
| "loss": 0.5922, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 30.5591677503251, | |
| "eval_accuracy": 0.8642099290677292, | |
| "eval_loss": 0.8136395215988159, | |
| "eval_runtime": 227.8537, | |
| "eval_samples_per_second": 89.68, | |
| "eval_steps_per_second": 2.804, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 30.721716514954487, | |
| "grad_norm": 2.54514479637146, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.5834, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 30.721716514954487, | |
| "eval_accuracy": 0.8647608401845014, | |
| "eval_loss": 0.8055408596992493, | |
| "eval_runtime": 222.2677, | |
| "eval_samples_per_second": 91.934, | |
| "eval_steps_per_second": 2.875, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 30.884265279583875, | |
| "grad_norm": 2.5504415035247803, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5851, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 30.884265279583875, | |
| "eval_accuracy": 0.8655900321339209, | |
| "eval_loss": 0.8102414011955261, | |
| "eval_runtime": 222.3833, | |
| "eval_samples_per_second": 91.886, | |
| "eval_steps_per_second": 2.873, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 31.046814044213264, | |
| "grad_norm": 2.5223162174224854, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.5858, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 31.046814044213264, | |
| "eval_accuracy": 0.8648693759071118, | |
| "eval_loss": 0.8096651434898376, | |
| "eval_runtime": 222.4149, | |
| "eval_samples_per_second": 91.873, | |
| "eval_steps_per_second": 2.873, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 31.209362808842652, | |
| "grad_norm": 2.704538583755493, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.5854, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 31.209362808842652, | |
| "eval_accuracy": 0.8643289659912363, | |
| "eval_loss": 0.8139033317565918, | |
| "eval_runtime": 222.3309, | |
| "eval_samples_per_second": 91.908, | |
| "eval_steps_per_second": 2.874, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 31.37191157347204, | |
| "grad_norm": 2.3286654949188232, | |
| "learning_rate": 3.5000000000000004e-06, | |
| "loss": 0.5809, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 31.37191157347204, | |
| "eval_accuracy": 0.8655915682628923, | |
| "eval_loss": 0.8057557940483093, | |
| "eval_runtime": 222.7618, | |
| "eval_samples_per_second": 91.73, | |
| "eval_steps_per_second": 2.869, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 31.53446033810143, | |
| "grad_norm": 2.572178363800049, | |
| "learning_rate": 3e-06, | |
| "loss": 0.5846, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 31.53446033810143, | |
| "eval_accuracy": 0.8653023190042686, | |
| "eval_loss": 0.8064507842063904, | |
| "eval_runtime": 223.3613, | |
| "eval_samples_per_second": 91.484, | |
| "eval_steps_per_second": 2.861, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 31.69700910273082, | |
| "grad_norm": 2.4184162616729736, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.585, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 31.69700910273082, | |
| "eval_accuracy": 0.8657269498564959, | |
| "eval_loss": 0.7961297631263733, | |
| "eval_runtime": 225.1117, | |
| "eval_samples_per_second": 90.773, | |
| "eval_steps_per_second": 2.839, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 31.85955786736021, | |
| "grad_norm": 2.121875047683716, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.588, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 31.85955786736021, | |
| "eval_accuracy": 0.8654759221390312, | |
| "eval_loss": 0.7994445562362671, | |
| "eval_runtime": 222.5184, | |
| "eval_samples_per_second": 91.831, | |
| "eval_steps_per_second": 2.872, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 32.022106631989594, | |
| "grad_norm": 2.2691831588745117, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.58, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 32.022106631989594, | |
| "eval_accuracy": 0.8651082523885062, | |
| "eval_loss": 0.8115944266319275, | |
| "eval_runtime": 222.6443, | |
| "eval_samples_per_second": 91.779, | |
| "eval_steps_per_second": 2.87, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 32.184655396618986, | |
| "grad_norm": 2.047391891479492, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.5776, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 32.184655396618986, | |
| "eval_accuracy": 0.8659600453633429, | |
| "eval_loss": 0.8002920150756836, | |
| "eval_runtime": 223.1406, | |
| "eval_samples_per_second": 91.575, | |
| "eval_steps_per_second": 2.864, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 32.34720416124838, | |
| "grad_norm": 2.4214348793029785, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.581, | |
| "step": 99500 | |
| }, | |
| { | |
| "epoch": 32.34720416124838, | |
| "eval_accuracy": 0.8655375271936185, | |
| "eval_loss": 0.7950243949890137, | |
| "eval_runtime": 240.505, | |
| "eval_samples_per_second": 84.963, | |
| "eval_steps_per_second": 2.657, | |
| "step": 99500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 33, | |
| "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": 8.401908931210772e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |