| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9995988768551946, | |
| "eval_steps": 100, | |
| "global_step": 623, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016044925792218213, | |
| "grad_norm": 0.07964161388964953, | |
| "learning_rate": 1.5873015873015873e-06, | |
| "loss": 1.478, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.032089851584436425, | |
| "grad_norm": 0.08446800204864474, | |
| "learning_rate": 3.1746031746031746e-06, | |
| "loss": 1.4734, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.048134777376654635, | |
| "grad_norm": 0.0880665470615365, | |
| "learning_rate": 4.761904761904762e-06, | |
| "loss": 1.4686, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06417970316887285, | |
| "grad_norm": 0.11036773606126081, | |
| "learning_rate": 6.349206349206349e-06, | |
| "loss": 1.4799, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08022462896109106, | |
| "grad_norm": 0.1984820440725813, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 1.4712, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09626955475330927, | |
| "grad_norm": 0.3636876671405763, | |
| "learning_rate": 9.523809523809525e-06, | |
| "loss": 1.4479, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11231448054552748, | |
| "grad_norm": 0.33961243491902277, | |
| "learning_rate": 9.996145181203616e-06, | |
| "loss": 1.3449, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1283594063377457, | |
| "grad_norm": 0.11623936014320749, | |
| "learning_rate": 9.977278743495434e-06, | |
| "loss": 1.3048, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1444043321299639, | |
| "grad_norm": 0.09934569363354419, | |
| "learning_rate": 9.942751945444437e-06, | |
| "loss": 1.2392, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.16044925792218212, | |
| "grad_norm": 0.0749779795039777, | |
| "learning_rate": 9.892673421130979e-06, | |
| "loss": 1.2155, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16044925792218212, | |
| "eval_loss": 1.2045714855194092, | |
| "eval_runtime": 36.9278, | |
| "eval_samples_per_second": 5.47, | |
| "eval_steps_per_second": 1.381, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.17649418371440032, | |
| "grad_norm": 0.06497134561472019, | |
| "learning_rate": 9.827200736119815e-06, | |
| "loss": 1.2129, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.19253910950661854, | |
| "grad_norm": 0.06149214025937759, | |
| "learning_rate": 9.746539891700558e-06, | |
| "loss": 1.1695, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.20858403529883673, | |
| "grad_norm": 0.05334876254417194, | |
| "learning_rate": 9.650944676731383e-06, | |
| "loss": 1.162, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.22462896109105496, | |
| "grad_norm": 0.04556668435804944, | |
| "learning_rate": 9.540715869125407e-06, | |
| "loss": 1.1491, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.24067388688327315, | |
| "grad_norm": 0.04356126220951498, | |
| "learning_rate": 9.416200289492092e-06, | |
| "loss": 1.1358, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2567188126754914, | |
| "grad_norm": 0.03723127058817997, | |
| "learning_rate": 9.27778970991129e-06, | |
| "loss": 1.1338, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.27276373846770957, | |
| "grad_norm": 0.03598344746987169, | |
| "learning_rate": 9.125919621273348e-06, | |
| "loss": 1.1333, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2888086642599278, | |
| "grad_norm": 0.03958997125425865, | |
| "learning_rate": 8.961067863063638e-06, | |
| "loss": 1.1382, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.304853590052146, | |
| "grad_norm": 0.033584445768925664, | |
| "learning_rate": 8.783753119902766e-06, | |
| "loss": 1.1329, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.32089851584436424, | |
| "grad_norm": 0.0370694089463755, | |
| "learning_rate": 8.594533289572852e-06, | |
| "loss": 1.1392, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.32089851584436424, | |
| "eval_loss": 1.1237837076187134, | |
| "eval_runtime": 36.9468, | |
| "eval_samples_per_second": 5.467, | |
| "eval_steps_per_second": 1.38, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3369434416365824, | |
| "grad_norm": 0.031404112904854176, | |
| "learning_rate": 8.39400372766471e-06, | |
| "loss": 1.1338, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.35298836742880063, | |
| "grad_norm": 0.03068751780314738, | |
| "learning_rate": 8.182795374368893e-06, | |
| "loss": 1.1313, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.36903329322101885, | |
| "grad_norm": 0.030467037926615172, | |
| "learning_rate": 7.961572769304437e-06, | |
| "loss": 1.1247, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3850782190132371, | |
| "grad_norm": 0.030773507709064975, | |
| "learning_rate": 7.731031960631354e-06, | |
| "loss": 1.1318, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4011231448054553, | |
| "grad_norm": 0.033328095202181884, | |
| "learning_rate": 7.491898315025615e-06, | |
| "loss": 1.1249, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.41716807059767347, | |
| "grad_norm": 0.03128616232047187, | |
| "learning_rate": 7.244924235407224e-06, | |
| "loss": 1.1162, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.4332129963898917, | |
| "grad_norm": 0.031188168038119562, | |
| "learning_rate": 6.990886793602268e-06, | |
| "loss": 1.1133, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.4492579221821099, | |
| "grad_norm": 0.03333159876932075, | |
| "learning_rate": 6.730585285387465e-06, | |
| "loss": 1.1056, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.46530284797432814, | |
| "grad_norm": 0.0316593444984735, | |
| "learning_rate": 6.464838715609945e-06, | |
| "loss": 1.1266, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.4813477737665463, | |
| "grad_norm": 0.03167501381379994, | |
| "learning_rate": 6.194483221294989e-06, | |
| "loss": 1.1181, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4813477737665463, | |
| "eval_loss": 1.1140353679656982, | |
| "eval_runtime": 36.7217, | |
| "eval_samples_per_second": 5.501, | |
| "eval_steps_per_second": 1.389, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.49739269955876453, | |
| "grad_norm": 0.03328370476401871, | |
| "learning_rate": 5.920369440849609e-06, | |
| "loss": 1.0962, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5134376253509828, | |
| "grad_norm": 0.03447585267125308, | |
| "learning_rate": 5.643359837639419e-06, | |
| "loss": 1.1171, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.529482551143201, | |
| "grad_norm": 0.03781496580664963, | |
| "learning_rate": 5.3643259863598015e-06, | |
| "loss": 1.1091, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5455274769354191, | |
| "grad_norm": 0.03172320948571133, | |
| "learning_rate": 5.084145830739462e-06, | |
| "loss": 1.1017, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5615724027276374, | |
| "grad_norm": 0.03412876166847143, | |
| "learning_rate": 4.803700921204659e-06, | |
| "loss": 1.1169, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5776173285198556, | |
| "grad_norm": 0.03375281781263732, | |
| "learning_rate": 4.5238736411954075e-06, | |
| "loss": 1.0998, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5936622543120738, | |
| "grad_norm": 0.03717435838572482, | |
| "learning_rate": 4.245544430860743e-06, | |
| "loss": 1.1117, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.609707180104292, | |
| "grad_norm": 0.029559551388140908, | |
| "learning_rate": 3.969589016868269e-06, | |
| "loss": 1.1283, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6257521058965102, | |
| "grad_norm": 0.04200189123455742, | |
| "learning_rate": 3.6968756570440735e-06, | |
| "loss": 1.1228, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6417970316887285, | |
| "grad_norm": 0.03378722314455019, | |
| "learning_rate": 3.42826240851239e-06, | |
| "loss": 1.1252, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6417970316887285, | |
| "eval_loss": 1.1096988916397095, | |
| "eval_runtime": 36.792, | |
| "eval_samples_per_second": 5.49, | |
| "eval_steps_per_second": 1.386, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6578419574809466, | |
| "grad_norm": 0.03734682816128997, | |
| "learning_rate": 3.1645944279304296e-06, | |
| "loss": 1.1085, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.6738868832731648, | |
| "grad_norm": 0.06074049582614517, | |
| "learning_rate": 2.906701312312861e-06, | |
| "loss": 1.0973, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6899318090653831, | |
| "grad_norm": 0.03310219959096242, | |
| "learning_rate": 2.6553944888126772e-06, | |
| "loss": 1.0965, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.7059767348576013, | |
| "grad_norm": 0.036021492275449166, | |
| "learning_rate": 2.4114646616711844e-06, | |
| "loss": 1.1028, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.7220216606498195, | |
| "grad_norm": 0.03370764681001696, | |
| "learning_rate": 2.175679324369913e-06, | |
| "loss": 1.1094, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7380665864420377, | |
| "grad_norm": 0.032620569117429374, | |
| "learning_rate": 1.948780344812181e-06, | |
| "loss": 1.103, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.7541115122342559, | |
| "grad_norm": 0.03518623684780172, | |
| "learning_rate": 1.7314816311322219e-06, | |
| "loss": 1.1019, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.7701564380264742, | |
| "grad_norm": 0.033154472751256124, | |
| "learning_rate": 1.5244668854760459e-06, | |
| "loss": 1.1025, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7862013638186923, | |
| "grad_norm": 0.03282196969531501, | |
| "learning_rate": 1.3283874528215735e-06, | |
| "loss": 1.0912, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8022462896109106, | |
| "grad_norm": 0.03261545464797978, | |
| "learning_rate": 1.143860271606333e-06, | |
| "loss": 1.1199, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8022462896109106, | |
| "eval_loss": 1.1079450845718384, | |
| "eval_runtime": 37.0687, | |
| "eval_samples_per_second": 5.449, | |
| "eval_steps_per_second": 1.376, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8182912154031288, | |
| "grad_norm": 0.03521866325178647, | |
| "learning_rate": 9.714659326109138e-07, | |
| "loss": 1.1117, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.8343361411953469, | |
| "grad_norm": 0.03534230366225303, | |
| "learning_rate": 8.117468522055578e-07, | |
| "loss": 1.1197, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.8503810669875652, | |
| "grad_norm": 0.03192663388371955, | |
| "learning_rate": 6.652055657075845e-07, | |
| "loss": 1.1212, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.8664259927797834, | |
| "grad_norm": 0.03381094832008424, | |
| "learning_rate": 5.323031462193757e-07, | |
| "loss": 1.1147, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.8824709185720016, | |
| "grad_norm": 0.0312520726762653, | |
| "learning_rate": 4.134577539217965e-07, | |
| "loss": 1.1094, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.8985158443642198, | |
| "grad_norm": 0.031196657419516906, | |
| "learning_rate": 3.0904332038757977e-07, | |
| "loss": 1.1008, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.914560770156438, | |
| "grad_norm": 0.03159621090668423, | |
| "learning_rate": 2.1938837205424002e-07, | |
| "loss": 1.1146, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.9306056959486563, | |
| "grad_norm": 0.03478277084648844, | |
| "learning_rate": 1.4477499655837278e-07, | |
| "loss": 1.1122, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.9466506217408744, | |
| "grad_norm": 0.03684852374687432, | |
| "learning_rate": 8.543795518357767e-08, | |
| "loss": 1.1185, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.9626955475330926, | |
| "grad_norm": 0.03649848908454171, | |
| "learning_rate": 4.15639442146093e-08, | |
| "loss": 1.1104, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9626955475330926, | |
| "eval_loss": 1.107527494430542, | |
| "eval_runtime": 36.8485, | |
| "eval_samples_per_second": 5.482, | |
| "eval_steps_per_second": 1.384, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9787404733253109, | |
| "grad_norm": 0.032198270473962474, | |
| "learning_rate": 1.3291007521799015e-08, | |
| "loss": 1.1028, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.9947853991175291, | |
| "grad_norm": 0.0365869343105355, | |
| "learning_rate": 7.081022239591173e-10, | |
| "loss": 1.1179, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.9995988768551946, | |
| "step": 623, | |
| "total_flos": 6308174864842752.0, | |
| "train_loss": 1.1638640125146074, | |
| "train_runtime": 6246.6188, | |
| "train_samples_per_second": 3.192, | |
| "train_steps_per_second": 0.1 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 623, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6308174864842752.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |