| { | |
| "best_metric": 0.9574718836285272, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/bert/bert-base-finetuned-pos-ud-Arabic-PADT/checkpoint-13000", | |
| "epoch": 78.94736842105263, | |
| "global_step": 15000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 1.2456, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 4.966778523489933e-05, | |
| "loss": 0.3317, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.933221476510068e-05, | |
| "loss": 0.2403, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.8996644295302016e-05, | |
| "loss": 0.2052, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 4.8661073825503355e-05, | |
| "loss": 0.171, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_accuracy": 0.9443396861741531, | |
| "eval_loss": 0.18388177454471588, | |
| "eval_runtime": 4.6604, | |
| "eval_samples_per_second": 195.048, | |
| "eval_steps_per_second": 24.461, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 4.83255033557047e-05, | |
| "loss": 0.1527, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 4.798993288590604e-05, | |
| "loss": 0.1387, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 4.765436241610739e-05, | |
| "loss": 0.1234, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 4.7318791946308726e-05, | |
| "loss": 0.1116, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 4.698322147651007e-05, | |
| "loss": 0.1049, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "eval_accuracy": 0.9493231867465822, | |
| "eval_loss": 0.18382076919078827, | |
| "eval_runtime": 4.6662, | |
| "eval_samples_per_second": 194.805, | |
| "eval_steps_per_second": 24.431, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 4.664765100671141e-05, | |
| "loss": 0.0932, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 4.631208053691276e-05, | |
| "loss": 0.0843, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 4.5976510067114097e-05, | |
| "loss": 0.0773, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 4.564093959731544e-05, | |
| "loss": 0.0664, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 4.5305369127516775e-05, | |
| "loss": 0.0678, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "eval_accuracy": 0.9527914337665836, | |
| "eval_loss": 0.18106327950954437, | |
| "eval_runtime": 4.676, | |
| "eval_samples_per_second": 194.398, | |
| "eval_steps_per_second": 24.38, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 4.496979865771812e-05, | |
| "loss": 0.0566, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 4.463422818791946e-05, | |
| "loss": 0.0567, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 4.4298657718120806e-05, | |
| "loss": 0.0443, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4.3963087248322146e-05, | |
| "loss": 0.0465, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "learning_rate": 4.362751677852349e-05, | |
| "loss": 0.0359, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "eval_accuracy": 0.9525893999595932, | |
| "eval_loss": 0.21346914768218994, | |
| "eval_runtime": 4.6791, | |
| "eval_samples_per_second": 194.267, | |
| "eval_steps_per_second": 24.364, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 4.329194630872484e-05, | |
| "loss": 0.0388, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "learning_rate": 4.295637583892618e-05, | |
| "loss": 0.0327, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 12.11, | |
| "learning_rate": 4.262080536912752e-05, | |
| "loss": 0.0319, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 12.63, | |
| "learning_rate": 4.228523489932886e-05, | |
| "loss": 0.028, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 4.194966442953021e-05, | |
| "loss": 0.0275, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "eval_accuracy": 0.9520843154421174, | |
| "eval_loss": 0.2462630569934845, | |
| "eval_runtime": 4.6736, | |
| "eval_samples_per_second": 194.499, | |
| "eval_steps_per_second": 24.393, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "learning_rate": 4.161409395973155e-05, | |
| "loss": 0.023, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 4.127852348993289e-05, | |
| "loss": 0.0241, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 4.0942953020134226e-05, | |
| "loss": 0.0214, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 15.26, | |
| "learning_rate": 4.060738255033557e-05, | |
| "loss": 0.0191, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 4.027181208053691e-05, | |
| "loss": 0.0188, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "eval_accuracy": 0.9528587783689138, | |
| "eval_loss": 0.24228344857692719, | |
| "eval_runtime": 4.6621, | |
| "eval_samples_per_second": 194.978, | |
| "eval_steps_per_second": 24.453, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 16.32, | |
| "learning_rate": 3.993624161073826e-05, | |
| "loss": 0.0166, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 3.96006711409396e-05, | |
| "loss": 0.0183, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 17.37, | |
| "learning_rate": 3.926510067114094e-05, | |
| "loss": 0.0155, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 3.892953020134228e-05, | |
| "loss": 0.0156, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "learning_rate": 3.859395973154363e-05, | |
| "loss": 0.0133, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "eval_accuracy": 0.95349855209105, | |
| "eval_loss": 0.26400476694107056, | |
| "eval_runtime": 4.6713, | |
| "eval_samples_per_second": 194.593, | |
| "eval_steps_per_second": 24.404, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 18.95, | |
| "learning_rate": 3.825838926174497e-05, | |
| "loss": 0.0142, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 3.7922818791946313e-05, | |
| "loss": 0.0122, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 3.758724832214765e-05, | |
| "loss": 0.0131, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 20.53, | |
| "learning_rate": 3.725167785234899e-05, | |
| "loss": 0.0115, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "learning_rate": 3.691610738255034e-05, | |
| "loss": 0.0127, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "eval_accuracy": 0.9521853323456125, | |
| "eval_loss": 0.27901268005371094, | |
| "eval_runtime": 4.6772, | |
| "eval_samples_per_second": 194.348, | |
| "eval_steps_per_second": 24.374, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 21.58, | |
| "learning_rate": 3.658053691275168e-05, | |
| "loss": 0.0107, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 22.11, | |
| "learning_rate": 3.624496644295302e-05, | |
| "loss": 0.0112, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "learning_rate": 3.590939597315436e-05, | |
| "loss": 0.0097, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 23.16, | |
| "learning_rate": 3.557382550335571e-05, | |
| "loss": 0.0101, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "learning_rate": 3.523825503355705e-05, | |
| "loss": 0.0091, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "eval_accuracy": 0.9530608121759041, | |
| "eval_loss": 0.2944227457046509, | |
| "eval_runtime": 4.6644, | |
| "eval_samples_per_second": 194.881, | |
| "eval_steps_per_second": 24.44, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 24.21, | |
| "learning_rate": 3.4902684563758394e-05, | |
| "loss": 0.0097, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 24.74, | |
| "learning_rate": 3.456711409395973e-05, | |
| "loss": 0.0075, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 25.26, | |
| "learning_rate": 3.423154362416108e-05, | |
| "loss": 0.0086, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "learning_rate": 3.389597315436242e-05, | |
| "loss": 0.0075, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 3.356040268456376e-05, | |
| "loss": 0.0082, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "eval_accuracy": 0.9545087211260017, | |
| "eval_loss": 0.28701817989349365, | |
| "eval_runtime": 4.6855, | |
| "eval_samples_per_second": 194.002, | |
| "eval_steps_per_second": 24.33, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 26.84, | |
| "learning_rate": 3.32248322147651e-05, | |
| "loss": 0.008, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 27.37, | |
| "learning_rate": 3.288926174496644e-05, | |
| "loss": 0.0066, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 27.89, | |
| "learning_rate": 3.255369127516779e-05, | |
| "loss": 0.007, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 28.42, | |
| "learning_rate": 3.221812080536913e-05, | |
| "loss": 0.0065, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 28.95, | |
| "learning_rate": 3.1882550335570474e-05, | |
| "loss": 0.0071, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 28.95, | |
| "eval_accuracy": 0.9537005858980403, | |
| "eval_loss": 0.2949673533439636, | |
| "eval_runtime": 4.6645, | |
| "eval_samples_per_second": 194.878, | |
| "eval_steps_per_second": 24.44, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 29.47, | |
| "learning_rate": 3.1546979865771814e-05, | |
| "loss": 0.0058, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 3.121140939597316e-05, | |
| "loss": 0.0069, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 30.53, | |
| "learning_rate": 3.08758389261745e-05, | |
| "loss": 0.0057, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 31.05, | |
| "learning_rate": 3.0540268456375845e-05, | |
| "loss": 0.0062, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 31.58, | |
| "learning_rate": 3.0204697986577184e-05, | |
| "loss": 0.0059, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 31.58, | |
| "eval_accuracy": 0.9539699643073608, | |
| "eval_loss": 0.30248409509658813, | |
| "eval_runtime": 4.6764, | |
| "eval_samples_per_second": 194.38, | |
| "eval_steps_per_second": 24.378, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 32.11, | |
| "learning_rate": 2.9869127516778524e-05, | |
| "loss": 0.0066, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 32.63, | |
| "learning_rate": 2.9533557046979866e-05, | |
| "loss": 0.0055, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 33.16, | |
| "learning_rate": 2.919798657718121e-05, | |
| "loss": 0.0058, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 33.68, | |
| "learning_rate": 2.886241610738255e-05, | |
| "loss": 0.0054, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 34.21, | |
| "learning_rate": 2.8526845637583894e-05, | |
| "loss": 0.0054, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 34.21, | |
| "eval_accuracy": 0.95349855209105, | |
| "eval_loss": 0.3090747594833374, | |
| "eval_runtime": 4.706, | |
| "eval_samples_per_second": 193.157, | |
| "eval_steps_per_second": 24.224, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 34.74, | |
| "learning_rate": 2.8191275167785237e-05, | |
| "loss": 0.0045, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 35.26, | |
| "learning_rate": 2.785570469798658e-05, | |
| "loss": 0.0048, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 35.79, | |
| "learning_rate": 2.7520134228187922e-05, | |
| "loss": 0.0051, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 36.32, | |
| "learning_rate": 2.7184563758389265e-05, | |
| "loss": 0.0041, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 36.84, | |
| "learning_rate": 2.6848993288590607e-05, | |
| "loss": 0.0043, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 36.84, | |
| "eval_accuracy": 0.9523873661526029, | |
| "eval_loss": 0.3119761049747467, | |
| "eval_runtime": 4.6767, | |
| "eval_samples_per_second": 194.369, | |
| "eval_steps_per_second": 24.376, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 37.37, | |
| "learning_rate": 2.651342281879195e-05, | |
| "loss": 0.0043, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 37.89, | |
| "learning_rate": 2.6177852348993293e-05, | |
| "loss": 0.0035, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 38.42, | |
| "learning_rate": 2.584228187919463e-05, | |
| "loss": 0.0035, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 38.95, | |
| "learning_rate": 2.551006711409396e-05, | |
| "loss": 0.0039, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 39.47, | |
| "learning_rate": 2.5174496644295303e-05, | |
| "loss": 0.0036, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 39.47, | |
| "eval_accuracy": 0.9550138056434777, | |
| "eval_loss": 0.317449688911438, | |
| "eval_runtime": 4.671, | |
| "eval_samples_per_second": 194.604, | |
| "eval_steps_per_second": 24.406, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 2.4838926174496646e-05, | |
| "loss": 0.0034, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 40.53, | |
| "learning_rate": 2.4506711409395974e-05, | |
| "loss": 0.0041, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 41.05, | |
| "learning_rate": 2.4171140939597317e-05, | |
| "loss": 0.0044, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 41.58, | |
| "learning_rate": 2.383557046979866e-05, | |
| "loss": 0.0039, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 42.11, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.0032, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 42.11, | |
| "eval_accuracy": 0.9548791164388174, | |
| "eval_loss": 0.32017943263053894, | |
| "eval_runtime": 4.682, | |
| "eval_samples_per_second": 194.147, | |
| "eval_steps_per_second": 24.348, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 42.63, | |
| "learning_rate": 2.316442953020134e-05, | |
| "loss": 0.0032, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 43.16, | |
| "learning_rate": 2.2828859060402684e-05, | |
| "loss": 0.0038, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 43.68, | |
| "learning_rate": 2.2493288590604027e-05, | |
| "loss": 0.0029, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 44.21, | |
| "learning_rate": 2.215771812080537e-05, | |
| "loss": 0.003, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 44.74, | |
| "learning_rate": 2.1822147651006712e-05, | |
| "loss": 0.0037, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 44.74, | |
| "eval_accuracy": 0.9536332412957101, | |
| "eval_loss": 0.3339065611362457, | |
| "eval_runtime": 4.676, | |
| "eval_samples_per_second": 194.396, | |
| "eval_steps_per_second": 24.38, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 45.26, | |
| "learning_rate": 2.1486577181208055e-05, | |
| "loss": 0.0033, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 45.79, | |
| "learning_rate": 2.1151006711409397e-05, | |
| "loss": 0.0038, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 46.32, | |
| "learning_rate": 2.081543624161074e-05, | |
| "loss": 0.0027, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 46.84, | |
| "learning_rate": 2.0479865771812082e-05, | |
| "loss": 0.0029, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 47.37, | |
| "learning_rate": 2.0144295302013425e-05, | |
| "loss": 0.0022, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 47.37, | |
| "eval_accuracy": 0.9552495117516331, | |
| "eval_loss": 0.3310903012752533, | |
| "eval_runtime": 4.6747, | |
| "eval_samples_per_second": 194.451, | |
| "eval_steps_per_second": 24.387, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 47.89, | |
| "learning_rate": 1.9808724832214768e-05, | |
| "loss": 0.002, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 48.42, | |
| "learning_rate": 1.9473154362416107e-05, | |
| "loss": 0.0023, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 48.95, | |
| "learning_rate": 1.913758389261745e-05, | |
| "loss": 0.0024, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 49.47, | |
| "learning_rate": 1.8802013422818792e-05, | |
| "loss": 0.0017, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.8466442953020135e-05, | |
| "loss": 0.0019, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.9556535793656138, | |
| "eval_loss": 0.3342548906803131, | |
| "eval_runtime": 4.6676, | |
| "eval_samples_per_second": 194.747, | |
| "eval_steps_per_second": 24.424, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 50.53, | |
| "learning_rate": 1.8130872483221478e-05, | |
| "loss": 0.002, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 51.05, | |
| "learning_rate": 1.779530201342282e-05, | |
| "loss": 0.0018, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 51.58, | |
| "learning_rate": 1.745973154362416e-05, | |
| "loss": 0.0019, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 52.11, | |
| "learning_rate": 1.7124161073825502e-05, | |
| "loss": 0.0019, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 52.63, | |
| "learning_rate": 1.6788590604026845e-05, | |
| "loss": 0.0018, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 52.63, | |
| "eval_accuracy": 0.955720923967944, | |
| "eval_loss": 0.34407833218574524, | |
| "eval_runtime": 4.6926, | |
| "eval_samples_per_second": 193.709, | |
| "eval_steps_per_second": 24.294, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 53.16, | |
| "learning_rate": 1.6453020134228188e-05, | |
| "loss": 0.0021, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 53.68, | |
| "learning_rate": 1.6117449664429534e-05, | |
| "loss": 0.0018, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 54.21, | |
| "learning_rate": 1.5781879194630873e-05, | |
| "loss": 0.0017, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 54.74, | |
| "learning_rate": 1.5446308724832215e-05, | |
| "loss": 0.0015, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 55.26, | |
| "learning_rate": 1.5110738255033558e-05, | |
| "loss": 0.0018, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 55.26, | |
| "eval_accuracy": 0.9560913192807596, | |
| "eval_loss": 0.3533622622489929, | |
| "eval_runtime": 4.6805, | |
| "eval_samples_per_second": 194.209, | |
| "eval_steps_per_second": 24.356, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 55.79, | |
| "learning_rate": 1.47751677852349e-05, | |
| "loss": 0.0017, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 56.32, | |
| "learning_rate": 1.4439597315436243e-05, | |
| "loss": 0.0019, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 56.84, | |
| "learning_rate": 1.4104026845637586e-05, | |
| "loss": 0.001, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 57.37, | |
| "learning_rate": 1.3768456375838925e-05, | |
| "loss": 0.0011, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 57.89, | |
| "learning_rate": 1.3432885906040268e-05, | |
| "loss": 0.0013, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 57.89, | |
| "eval_accuracy": 0.9553505286551283, | |
| "eval_loss": 0.3556341230869293, | |
| "eval_runtime": 4.6828, | |
| "eval_samples_per_second": 194.116, | |
| "eval_steps_per_second": 24.345, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 58.42, | |
| "learning_rate": 1.309731543624161e-05, | |
| "loss": 0.0013, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 58.95, | |
| "learning_rate": 1.2761744966442953e-05, | |
| "loss": 0.0012, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 59.47, | |
| "learning_rate": 1.2426174496644296e-05, | |
| "loss": 0.001, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "learning_rate": 1.2090604026845639e-05, | |
| "loss": 0.0014, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 60.53, | |
| "learning_rate": 1.1755033557046981e-05, | |
| "loss": 0.0017, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 60.53, | |
| "eval_accuracy": 0.9556872516667789, | |
| "eval_loss": 0.3435748815536499, | |
| "eval_runtime": 4.6824, | |
| "eval_samples_per_second": 194.129, | |
| "eval_steps_per_second": 24.346, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 61.05, | |
| "learning_rate": 1.1419463087248322e-05, | |
| "loss": 0.0013, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 61.58, | |
| "learning_rate": 1.1083892617449665e-05, | |
| "loss": 0.0008, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 62.11, | |
| "learning_rate": 1.0748322147651008e-05, | |
| "loss": 0.0012, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 62.63, | |
| "learning_rate": 1.0412751677852349e-05, | |
| "loss": 0.0009, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 63.16, | |
| "learning_rate": 1.0077181208053691e-05, | |
| "loss": 0.0011, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 63.16, | |
| "eval_accuracy": 0.9564617145935753, | |
| "eval_loss": 0.3513157069683075, | |
| "eval_runtime": 4.6776, | |
| "eval_samples_per_second": 194.332, | |
| "eval_steps_per_second": 24.372, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 63.68, | |
| "learning_rate": 9.741610738255034e-06, | |
| "loss": 0.0009, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 64.21, | |
| "learning_rate": 9.406040268456376e-06, | |
| "loss": 0.0008, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 64.74, | |
| "learning_rate": 9.070469798657719e-06, | |
| "loss": 0.0009, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 65.26, | |
| "learning_rate": 8.734899328859062e-06, | |
| "loss": 0.0009, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 65.79, | |
| "learning_rate": 8.40268456375839e-06, | |
| "loss": 0.0009, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 65.79, | |
| "eval_accuracy": 0.9567647653040608, | |
| "eval_loss": 0.35964664816856384, | |
| "eval_runtime": 4.6795, | |
| "eval_samples_per_second": 194.25, | |
| "eval_steps_per_second": 24.361, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 66.32, | |
| "learning_rate": 8.067114093959733e-06, | |
| "loss": 0.0008, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 66.84, | |
| "learning_rate": 7.731543624161074e-06, | |
| "loss": 0.0007, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 67.37, | |
| "learning_rate": 7.395973154362416e-06, | |
| "loss": 0.0007, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 67.89, | |
| "learning_rate": 7.060402684563759e-06, | |
| "loss": 0.0008, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 68.42, | |
| "learning_rate": 6.724832214765101e-06, | |
| "loss": 0.0006, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 68.42, | |
| "eval_accuracy": 0.9574718836285272, | |
| "eval_loss": 0.36283352971076965, | |
| "eval_runtime": 4.6706, | |
| "eval_samples_per_second": 194.621, | |
| "eval_steps_per_second": 24.408, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 68.95, | |
| "learning_rate": 6.3892617449664435e-06, | |
| "loss": 0.0007, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 69.47, | |
| "learning_rate": 6.053691275167785e-06, | |
| "loss": 0.0004, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "learning_rate": 5.718120805369128e-06, | |
| "loss": 0.0007, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 70.53, | |
| "learning_rate": 5.38255033557047e-06, | |
| "loss": 0.0004, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 71.05, | |
| "learning_rate": 5.0469798657718124e-06, | |
| "loss": 0.0006, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 71.05, | |
| "eval_accuracy": 0.9566300760994007, | |
| "eval_loss": 0.3649657964706421, | |
| "eval_runtime": 4.689, | |
| "eval_samples_per_second": 193.859, | |
| "eval_steps_per_second": 24.312, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 71.58, | |
| "learning_rate": 4.711409395973155e-06, | |
| "loss": 0.0005, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 72.11, | |
| "learning_rate": 4.375838926174497e-06, | |
| "loss": 0.0005, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 72.63, | |
| "learning_rate": 4.040268456375839e-06, | |
| "loss": 0.0006, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 73.16, | |
| "learning_rate": 3.704697986577181e-06, | |
| "loss": 0.0006, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 73.68, | |
| "learning_rate": 3.369127516778524e-06, | |
| "loss": 0.0003, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 73.68, | |
| "eval_accuracy": 0.9569331268098862, | |
| "eval_loss": 0.3655867874622345, | |
| "eval_runtime": 4.6863, | |
| "eval_samples_per_second": 193.968, | |
| "eval_steps_per_second": 24.326, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 74.21, | |
| "learning_rate": 3.033557046979866e-06, | |
| "loss": 0.0004, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 74.74, | |
| "learning_rate": 2.6979865771812085e-06, | |
| "loss": 0.0004, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 75.26, | |
| "learning_rate": 2.3624161073825503e-06, | |
| "loss": 0.0005, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 75.79, | |
| "learning_rate": 2.026845637583893e-06, | |
| "loss": 0.0004, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 76.32, | |
| "learning_rate": 1.6912751677852347e-06, | |
| "loss": 0.0005, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 76.32, | |
| "eval_accuracy": 0.9568994545087212, | |
| "eval_loss": 0.3665154278278351, | |
| "eval_runtime": 4.6825, | |
| "eval_samples_per_second": 194.127, | |
| "eval_steps_per_second": 24.346, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 76.84, | |
| "learning_rate": 1.3557046979865772e-06, | |
| "loss": 0.0004, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 77.37, | |
| "learning_rate": 1.0201342281879194e-06, | |
| "loss": 0.0004, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 77.89, | |
| "learning_rate": 6.845637583892617e-07, | |
| "loss": 0.0004, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 78.42, | |
| "learning_rate": 3.48993288590604e-07, | |
| "loss": 0.0003, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 78.95, | |
| "learning_rate": 1.3422818791946309e-08, | |
| "loss": 0.0003, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 78.95, | |
| "eval_accuracy": 0.9568994545087212, | |
| "eval_loss": 0.3661660850048065, | |
| "eval_runtime": 4.6775, | |
| "eval_samples_per_second": 194.335, | |
| "eval_steps_per_second": 24.372, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 78.95, | |
| "step": 15000, | |
| "total_flos": 6.266876636395008e+16, | |
| "train_loss": 0.028036758323883016, | |
| "train_runtime": 3937.3195, | |
| "train_samples_per_second": 121.91, | |
| "train_steps_per_second": 3.81 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 79, | |
| "total_flos": 6.266876636395008e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |