| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 610, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16393442622950818, | |
| "grad_norm": 0.4943847358226776, | |
| "learning_rate": 0.00019986740898848306, | |
| "loss": 1.1081, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32786885245901637, | |
| "grad_norm": 0.44973108172416687, | |
| "learning_rate": 0.0001994699875614589, | |
| "loss": 0.9226, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4918032786885246, | |
| "grad_norm": 0.3362947404384613, | |
| "learning_rate": 0.00019880878960910772, | |
| "loss": 0.887, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6557377049180327, | |
| "grad_norm": 0.36153319478034973, | |
| "learning_rate": 0.0001978855685095358, | |
| "loss": 0.8615, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.819672131147541, | |
| "grad_norm": 0.36022189259529114, | |
| "learning_rate": 0.00019670277247913205, | |
| "loss": 0.8234, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9836065573770492, | |
| "grad_norm": 0.4029320478439331, | |
| "learning_rate": 0.00019526353808033825, | |
| "loss": 0.8149, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1475409836065573, | |
| "grad_norm": 0.3638781011104584, | |
| "learning_rate": 0.00019357168190404936, | |
| "loss": 0.799, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.3114754098360657, | |
| "grad_norm": 0.3549768924713135, | |
| "learning_rate": 0.0001916316904487005, | |
| "loss": 0.785, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.4754098360655736, | |
| "grad_norm": 0.3937470018863678, | |
| "learning_rate": 0.00018944870822287956, | |
| "loss": 0.7776, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.639344262295082, | |
| "grad_norm": 0.36256495118141174, | |
| "learning_rate": 0.00018702852410301554, | |
| "loss": 0.7756, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.8032786885245902, | |
| "grad_norm": 0.3645057678222656, | |
| "learning_rate": 0.00018437755598231856, | |
| "loss": 0.767, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.9672131147540983, | |
| "grad_norm": 0.3446103632450104, | |
| "learning_rate": 0.00018150283375168114, | |
| "loss": 0.7543, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.1311475409836067, | |
| "grad_norm": 0.4181569516658783, | |
| "learning_rate": 0.00017841198065767107, | |
| "loss": 0.7231, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.2950819672131146, | |
| "grad_norm": 0.40038439631462097, | |
| "learning_rate": 0.00017511319308705198, | |
| "loss": 0.7348, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.459016393442623, | |
| "grad_norm": 0.3984282612800598, | |
| "learning_rate": 0.00017161521883143934, | |
| "loss": 0.7255, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.6229508196721314, | |
| "grad_norm": 0.4186936318874359, | |
| "learning_rate": 0.00016792733388972932, | |
| "loss": 0.7263, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.7868852459016393, | |
| "grad_norm": 0.4004381597042084, | |
| "learning_rate": 0.00016405931786981755, | |
| "loss": 0.7271, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.9508196721311473, | |
| "grad_norm": 0.4597119390964508, | |
| "learning_rate": 0.00016002142805483685, | |
| "loss": 0.7125, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.1147540983606556, | |
| "grad_norm": 0.4180513322353363, | |
| "learning_rate": 0.00015582437220268647, | |
| "loss": 0.6925, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.278688524590164, | |
| "grad_norm": 0.4401358366012573, | |
| "learning_rate": 0.0001514792801509831, | |
| "loss": 0.6863, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.442622950819672, | |
| "grad_norm": 0.4415852427482605, | |
| "learning_rate": 0.000146997674302732, | |
| "loss": 0.6857, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.6065573770491803, | |
| "grad_norm": 0.4144597053527832, | |
| "learning_rate": 0.0001423914390709861, | |
| "loss": 0.6807, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.7704918032786887, | |
| "grad_norm": 0.4162183403968811, | |
| "learning_rate": 0.00013767278936351854, | |
| "loss": 0.6846, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.9344262295081966, | |
| "grad_norm": 0.42117443680763245, | |
| "learning_rate": 0.0001328542381910835, | |
| "loss": 0.6887, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.098360655737705, | |
| "grad_norm": 0.4591551125049591, | |
| "learning_rate": 0.00012794856348516095, | |
| "loss": 0.6543, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.262295081967213, | |
| "grad_norm": 0.4592267572879791, | |
| "learning_rate": 0.0001229687742131796, | |
| "loss": 0.6482, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.426229508196721, | |
| "grad_norm": 0.47369667887687683, | |
| "learning_rate": 0.00011792807588107357, | |
| "loss": 0.6494, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.590163934426229, | |
| "grad_norm": 0.45578595995903015, | |
| "learning_rate": 0.00011283983551465511, | |
| "loss": 0.6559, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.754098360655737, | |
| "grad_norm": 0.4742227792739868, | |
| "learning_rate": 0.00010771754621266466, | |
| "loss": 0.6462, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.918032786885246, | |
| "grad_norm": 0.48461949825286865, | |
| "learning_rate": 0.00010257479136549889, | |
| "loss": 0.6501, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.081967213114754, | |
| "grad_norm": 0.4797608554363251, | |
| "learning_rate": 9.742520863450115e-05, | |
| "loss": 0.6452, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.245901639344262, | |
| "grad_norm": 0.505832314491272, | |
| "learning_rate": 9.228245378733537e-05, | |
| "loss": 0.6178, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.409836065573771, | |
| "grad_norm": 0.5265078544616699, | |
| "learning_rate": 8.71601644853449e-05, | |
| "loss": 0.6177, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.573770491803279, | |
| "grad_norm": 0.5109356045722961, | |
| "learning_rate": 8.207192411892646e-05, | |
| "loss": 0.6225, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.737704918032787, | |
| "grad_norm": 0.5139452815055847, | |
| "learning_rate": 7.703122578682046e-05, | |
| "loss": 0.6229, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.901639344262295, | |
| "grad_norm": 0.5581080913543701, | |
| "learning_rate": 7.205143651483906e-05, | |
| "loss": 0.6189, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.065573770491803, | |
| "grad_norm": 0.4794338345527649, | |
| "learning_rate": 6.714576180891654e-05, | |
| "loss": 0.6087, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 6.229508196721311, | |
| "grad_norm": 0.5142917037010193, | |
| "learning_rate": 6.232721063648148e-05, | |
| "loss": 0.5977, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.39344262295082, | |
| "grad_norm": 0.547099769115448, | |
| "learning_rate": 5.7608560929013946e-05, | |
| "loss": 0.6058, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.557377049180328, | |
| "grad_norm": 0.5487104654312134, | |
| "learning_rate": 5.300232569726804e-05, | |
| "loss": 0.5939, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.721311475409836, | |
| "grad_norm": 0.5189688205718994, | |
| "learning_rate": 4.852071984901696e-05, | |
| "loss": 0.6015, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.885245901639344, | |
| "grad_norm": 0.5185168981552124, | |
| "learning_rate": 4.417562779731355e-05, | |
| "loss": 0.5949, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 7.049180327868853, | |
| "grad_norm": 0.5486496090888977, | |
| "learning_rate": 3.997857194516319e-05, | |
| "loss": 0.5957, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 7.213114754098361, | |
| "grad_norm": 0.5592466592788696, | |
| "learning_rate": 3.594068213018249e-05, | |
| "loss": 0.582, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.377049180327869, | |
| "grad_norm": 0.5590682625770569, | |
| "learning_rate": 3.207266611027069e-05, | |
| "loss": 0.5779, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.540983606557377, | |
| "grad_norm": 0.546293318271637, | |
| "learning_rate": 2.8384781168560693e-05, | |
| "loss": 0.5736, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.704918032786885, | |
| "grad_norm": 0.5725670456886292, | |
| "learning_rate": 2.4886806912948035e-05, | |
| "loss": 0.5825, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.868852459016393, | |
| "grad_norm": 0.574937105178833, | |
| "learning_rate": 2.1588019342328968e-05, | |
| "loss": 0.5843, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 8.032786885245901, | |
| "grad_norm": 0.5568664073944092, | |
| "learning_rate": 1.8497166248318876e-05, | |
| "loss": 0.5791, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 8.19672131147541, | |
| "grad_norm": 0.568524181842804, | |
| "learning_rate": 1.562244401768144e-05, | |
| "loss": 0.5673, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.360655737704919, | |
| "grad_norm": 0.5631764531135559, | |
| "learning_rate": 1.2971475896984475e-05, | |
| "loss": 0.5557, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.524590163934427, | |
| "grad_norm": 0.5955241322517395, | |
| "learning_rate": 1.0551291777120464e-05, | |
| "loss": 0.5741, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.688524590163935, | |
| "grad_norm": 0.5839523077011108, | |
| "learning_rate": 8.368309551299536e-06, | |
| "loss": 0.5674, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.852459016393443, | |
| "grad_norm": 0.5877330899238586, | |
| "learning_rate": 6.428318095950647e-06, | |
| "loss": 0.5819, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 9.01639344262295, | |
| "grad_norm": 0.5768188238143921, | |
| "learning_rate": 4.7364619196617495e-06, | |
| "loss": 0.5728, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 9.180327868852459, | |
| "grad_norm": 0.5796675682067871, | |
| "learning_rate": 3.2972275208679625e-06, | |
| "loss": 0.5682, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 9.344262295081966, | |
| "grad_norm": 0.6149064898490906, | |
| "learning_rate": 2.1144314904642195e-06, | |
| "loss": 0.5699, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 9.508196721311476, | |
| "grad_norm": 0.558674156665802, | |
| "learning_rate": 1.1912103908922945e-06, | |
| "loss": 0.5663, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.672131147540984, | |
| "grad_norm": 0.563441812992096, | |
| "learning_rate": 5.300124385410943e-07, | |
| "loss": 0.5626, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.836065573770492, | |
| "grad_norm": 0.5652551651000977, | |
| "learning_rate": 1.3259101151694708e-07, | |
| "loss": 0.5579, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.5645861625671387, | |
| "learning_rate": 0.0, | |
| "loss": 0.5631, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 610, | |
| "total_flos": 4.672887793385472e+16, | |
| "train_loss": 0.6668467552935491, | |
| "train_runtime": 2732.8065, | |
| "train_samples_per_second": 2.679, | |
| "train_steps_per_second": 0.223 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 610, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 4.672887793385472e+16, | |
| "train_batch_size": 3, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |