| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.5048908954100828, | |
| "eval_steps": 20, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 5e-07, | |
| "loss": 1.1046, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1e-06, | |
| "loss": 1.0664, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.44479243019924036, | |
| "eval_combined_score": 0.3403229798934559, | |
| "eval_f1": 0.2738667532127967, | |
| "eval_loss": 1.065091609954834, | |
| "eval_precision": 0.19784030596254612, | |
| "eval_recall": 0.44479243019924036, | |
| "eval_runtime": 49.6621, | |
| "eval_samples_per_second": 604.364, | |
| "eval_steps_per_second": 1.188, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.0672, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2e-06, | |
| "loss": 1.0423, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.5033650962883988, | |
| "eval_combined_score": 0.4633375988688383, | |
| "eval_f1": 0.4137746858205889, | |
| "eval_loss": 1.0188277959823608, | |
| "eval_precision": 0.43284551707796665, | |
| "eval_recall": 0.5033650962883988, | |
| "eval_runtime": 49.1985, | |
| "eval_samples_per_second": 610.06, | |
| "eval_steps_per_second": 1.199, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.027, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3e-06, | |
| "loss": 1.0137, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.5279203038581995, | |
| "eval_combined_score": 0.48433586826626607, | |
| "eval_f1": 0.4596383010419703, | |
| "eval_loss": 0.9871189594268799, | |
| "eval_precision": 0.421864564306695, | |
| "eval_recall": 0.5279203038581995, | |
| "eval_runtime": 49.0237, | |
| "eval_samples_per_second": 612.235, | |
| "eval_steps_per_second": 1.204, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.0005, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4e-06, | |
| "loss": 1.0027, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.5308189511561271, | |
| "eval_combined_score": 0.4866197693698143, | |
| "eval_f1": 0.46528335920671143, | |
| "eval_loss": 0.9888613224029541, | |
| "eval_precision": 0.4195578159602916, | |
| "eval_recall": 0.5308189511561271, | |
| "eval_runtime": 49.518, | |
| "eval_samples_per_second": 606.124, | |
| "eval_steps_per_second": 1.191, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.9937, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9914, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.5307856333710935, | |
| "eval_combined_score": 0.48644275806230897, | |
| "eval_f1": 0.4666072639999521, | |
| "eval_loss": 0.9762536883354187, | |
| "eval_precision": 0.4175925015070968, | |
| "eval_recall": 0.5307856333710935, | |
| "eval_runtime": 49.4629, | |
| "eval_samples_per_second": 606.798, | |
| "eval_steps_per_second": 1.193, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 5.5e-06, | |
| "loss": 0.9884, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 6e-06, | |
| "loss": 0.9826, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.5387819017791697, | |
| "eval_combined_score": 0.4944622623129249, | |
| "eval_f1": 0.47105988260831677, | |
| "eval_loss": 0.9713281989097595, | |
| "eval_precision": 0.4292253630850433, | |
| "eval_recall": 0.5387819017791697, | |
| "eval_runtime": 49.0682, | |
| "eval_samples_per_second": 611.68, | |
| "eval_steps_per_second": 1.202, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.5e-06, | |
| "loss": 0.9882, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 7e-06, | |
| "loss": 0.9788, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_accuracy": 0.5312520823615646, | |
| "eval_combined_score": 0.48712844254868687, | |
| "eval_f1": 0.4673598197970026, | |
| "eval_loss": 0.9766249656677246, | |
| "eval_precision": 0.41864978567461564, | |
| "eval_recall": 0.5312520823615646, | |
| "eval_runtime": 49.1859, | |
| "eval_samples_per_second": 610.215, | |
| "eval_steps_per_second": 1.2, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.9854, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 8e-06, | |
| "loss": 0.984, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.5398147531152129, | |
| "eval_combined_score": 0.49477075635813, | |
| "eval_f1": 0.4751263005883661, | |
| "eval_loss": 0.9589501619338989, | |
| "eval_precision": 0.4243272186137281, | |
| "eval_recall": 0.5398147531152129, | |
| "eval_runtime": 49.0847, | |
| "eval_samples_per_second": 611.473, | |
| "eval_steps_per_second": 1.202, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.9795, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9e-06, | |
| "loss": 0.9694, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.5423469047777704, | |
| "eval_combined_score": 0.49718028497336725, | |
| "eval_f1": 0.4771728160733735, | |
| "eval_loss": 0.953514814376831, | |
| "eval_precision": 0.42685451426455484, | |
| "eval_recall": 0.5423469047777704, | |
| "eval_runtime": 49.0032, | |
| "eval_samples_per_second": 612.491, | |
| "eval_steps_per_second": 1.204, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.9757, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9676, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.567201972412874, | |
| "eval_combined_score": 0.5200447629299639, | |
| "eval_f1": 0.49911443945569844, | |
| "eval_loss": 0.9273685812950134, | |
| "eval_precision": 0.44666066743840943, | |
| "eval_recall": 0.567201972412874, | |
| "eval_runtime": 49.1912, | |
| "eval_samples_per_second": 610.15, | |
| "eval_steps_per_second": 1.199, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.0500000000000001e-05, | |
| "loss": 0.9708, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.1e-05, | |
| "loss": 0.9753, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.573598987139335, | |
| "eval_combined_score": 0.5278568904198743, | |
| "eval_f1": 0.5026102591641352, | |
| "eval_loss": 0.912590503692627, | |
| "eval_precision": 0.4616193282366919, | |
| "eval_recall": 0.573598987139335, | |
| "eval_runtime": 49.2154, | |
| "eval_samples_per_second": 609.849, | |
| "eval_steps_per_second": 1.199, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.15e-05, | |
| "loss": 0.9586, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.9557, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.5759645498767242, | |
| "eval_combined_score": 0.5280177627115059, | |
| "eval_f1": 0.5069205876947673, | |
| "eval_loss": 0.9052607417106628, | |
| "eval_precision": 0.45322136339780783, | |
| "eval_recall": 0.5759645498767242, | |
| "eval_runtime": 49.1399, | |
| "eval_samples_per_second": 610.787, | |
| "eval_steps_per_second": 1.201, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.9518, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.3e-05, | |
| "loss": 0.9508, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_accuracy": 0.5766975411474645, | |
| "eval_combined_score": 0.534067292009198, | |
| "eval_f1": 0.5018221478654629, | |
| "eval_loss": 0.9178985953330994, | |
| "eval_precision": 0.4810519378764, | |
| "eval_recall": 0.5766975411474645, | |
| "eval_runtime": 49.2092, | |
| "eval_samples_per_second": 609.927, | |
| "eval_steps_per_second": 1.199, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.35e-05, | |
| "loss": 0.9405, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.4e-05, | |
| "loss": 0.9355, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.58915839275005, | |
| "eval_combined_score": 0.5407092882328752, | |
| "eval_f1": 0.5183161389995846, | |
| "eval_loss": 0.8937407732009888, | |
| "eval_precision": 0.46620422843181647, | |
| "eval_recall": 0.58915839275005, | |
| "eval_runtime": 49.2523, | |
| "eval_samples_per_second": 609.393, | |
| "eval_steps_per_second": 1.198, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.4500000000000002e-05, | |
| "loss": 0.9235, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.613013926834144, | |
| "eval_combined_score": 0.5923759012239074, | |
| "eval_f1": 0.5579282363395321, | |
| "eval_loss": 0.8468813300132751, | |
| "eval_precision": 0.5855475148878095, | |
| "eval_recall": 0.613013926834144, | |
| "eval_runtime": 49.1896, | |
| "eval_samples_per_second": 610.17, | |
| "eval_steps_per_second": 1.199, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.8857, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.6e-05, | |
| "loss": 0.993, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.6046511627906976, | |
| "eval_combined_score": 0.590510828847508, | |
| "eval_f1": 0.5352474275558012, | |
| "eval_loss": 0.8615403771400452, | |
| "eval_precision": 0.6174935622528357, | |
| "eval_recall": 0.6046511627906976, | |
| "eval_runtime": 49.37, | |
| "eval_samples_per_second": 607.94, | |
| "eval_steps_per_second": 1.195, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.8418, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.8527, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_accuracy": 0.6439328313453722, | |
| "eval_combined_score": 0.6332615383344228, | |
| "eval_f1": 0.6212200513859825, | |
| "eval_loss": 0.7896137237548828, | |
| "eval_precision": 0.6239604392609642, | |
| "eval_recall": 0.6439328313453722, | |
| "eval_runtime": 49.1146, | |
| "eval_samples_per_second": 611.102, | |
| "eval_steps_per_second": 1.201, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 0.8982, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.966, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_accuracy": 0.5315852602119011, | |
| "eval_combined_score": 0.5500800253024862, | |
| "eval_f1": 0.4943799031015135, | |
| "eval_loss": 1.0123510360717773, | |
| "eval_precision": 0.642769677684629, | |
| "eval_recall": 0.5315852602119011, | |
| "eval_runtime": 49.1941, | |
| "eval_samples_per_second": 610.114, | |
| "eval_steps_per_second": 1.199, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.9413, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.8441, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.6488638635303525, | |
| "eval_combined_score": 0.648329789222037, | |
| "eval_f1": 0.6371420074616733, | |
| "eval_loss": 0.791083574295044, | |
| "eval_precision": 0.6584494223657698, | |
| "eval_recall": 0.6488638635303525, | |
| "eval_runtime": 49.3484, | |
| "eval_samples_per_second": 608.207, | |
| "eval_steps_per_second": 1.196, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.95e-05, | |
| "loss": 0.8223, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8226, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.6699873392416872, | |
| "eval_combined_score": 0.6621103162794577, | |
| "eval_f1": 0.6459453754294268, | |
| "eval_loss": 0.7472424507141113, | |
| "eval_precision": 0.6625212112050299, | |
| "eval_recall": 0.6699873392416872, | |
| "eval_runtime": 49.3153, | |
| "eval_samples_per_second": 608.614, | |
| "eval_steps_per_second": 1.196, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.7924, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.1000000000000002e-05, | |
| "loss": 0.7948, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_accuracy": 0.6580595721996402, | |
| "eval_combined_score": 0.6513132521613654, | |
| "eval_f1": 0.6095365334747936, | |
| "eval_loss": 0.7663838863372803, | |
| "eval_precision": 0.6795973307713873, | |
| "eval_recall": 0.6580595721996402, | |
| "eval_runtime": 49.2143, | |
| "eval_samples_per_second": 609.863, | |
| "eval_steps_per_second": 1.199, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 2.1499999999999997e-05, | |
| "loss": 0.776, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.2e-05, | |
| "loss": 0.7428, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_accuracy": 0.6991737189311654, | |
| "eval_combined_score": 0.6905198116505921, | |
| "eval_f1": 0.653333567856404, | |
| "eval_loss": 0.6993714570999146, | |
| "eval_precision": 0.7103982408836337, | |
| "eval_recall": 0.6991737189311654, | |
| "eval_runtime": 49.1465, | |
| "eval_samples_per_second": 610.704, | |
| "eval_steps_per_second": 1.2, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 0.7512, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.3e-05, | |
| "loss": 0.7109, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_accuracy": 0.7283600986206437, | |
| "eval_combined_score": 0.7200537958840363, | |
| "eval_f1": 0.6890059409106147, | |
| "eval_loss": 0.6510820984840393, | |
| "eval_precision": 0.7344890453842432, | |
| "eval_recall": 0.7283600986206437, | |
| "eval_runtime": 49.0446, | |
| "eval_samples_per_second": 611.974, | |
| "eval_steps_per_second": 1.203, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.3500000000000002e-05, | |
| "loss": 0.701, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.6882, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_accuracy": 0.7576797494502565, | |
| "eval_combined_score": 0.7514829004826655, | |
| "eval_f1": 0.7336051419487455, | |
| "eval_loss": 0.5987845063209534, | |
| "eval_precision": 0.7569669610814035, | |
| "eval_recall": 0.7576797494502565, | |
| "eval_runtime": 49.3939, | |
| "eval_samples_per_second": 607.646, | |
| "eval_steps_per_second": 1.194, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 2.4500000000000003e-05, | |
| "loss": 0.651, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7296, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_accuracy": 0.7564136736189778, | |
| "eval_combined_score": 0.7674838523650971, | |
| "eval_f1": 0.766902713706533, | |
| "eval_loss": 0.5992804765701294, | |
| "eval_precision": 0.7902053485158999, | |
| "eval_recall": 0.7564136736189778, | |
| "eval_runtime": 49.1622, | |
| "eval_samples_per_second": 610.509, | |
| "eval_steps_per_second": 1.2, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1660, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "total_flos": 6.735493003102618e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |