| { | |
| "best_global_step": 2000, | |
| "best_metric": 1.2713440656661987, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 1000, | |
| "global_step": 2906, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017205781142463867, | |
| "grad_norm": 0.8632408976554871, | |
| "learning_rate": 2.784090909090909e-05, | |
| "loss": 1.268, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.034411562284927734, | |
| "grad_norm": 0.8244401812553406, | |
| "learning_rate": 4.9804826117814056e-05, | |
| "loss": 1.2135, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.051617343427391604, | |
| "grad_norm": 1.049148678779602, | |
| "learning_rate": 4.891767210787793e-05, | |
| "loss": 1.1783, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06882312456985547, | |
| "grad_norm": 0.8507347106933594, | |
| "learning_rate": 4.8030518097941806e-05, | |
| "loss": 1.1688, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08602890571231935, | |
| "grad_norm": 0.8333495259284973, | |
| "learning_rate": 4.714336408800568e-05, | |
| "loss": 1.1632, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.10323468685478321, | |
| "grad_norm": 0.8183310627937317, | |
| "learning_rate": 4.6256210078069556e-05, | |
| "loss": 1.1241, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12044046799724707, | |
| "grad_norm": 0.8449575901031494, | |
| "learning_rate": 4.536905606813343e-05, | |
| "loss": 1.1488, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.13764624913971094, | |
| "grad_norm": 0.8357868194580078, | |
| "learning_rate": 4.448190205819731e-05, | |
| "loss": 1.1488, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1548520302821748, | |
| "grad_norm": 0.922451913356781, | |
| "learning_rate": 4.359474804826118e-05, | |
| "loss": 1.1242, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1720578114246387, | |
| "grad_norm": 0.9548965096473694, | |
| "learning_rate": 4.270759403832506e-05, | |
| "loss": 1.1569, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.18926359256710254, | |
| "grad_norm": 0.8528347611427307, | |
| "learning_rate": 4.182044002838893e-05, | |
| "loss": 1.1645, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.20646937370956642, | |
| "grad_norm": 0.9261412620544434, | |
| "learning_rate": 4.093328601845281e-05, | |
| "loss": 1.1587, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2236751548520303, | |
| "grad_norm": 0.8746649622917175, | |
| "learning_rate": 4.0046132008516676e-05, | |
| "loss": 1.1441, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.24088093599449414, | |
| "grad_norm": 0.8994395732879639, | |
| "learning_rate": 3.915897799858055e-05, | |
| "loss": 1.1248, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.258086717136958, | |
| "grad_norm": 0.8658247590065002, | |
| "learning_rate": 3.8271823988644426e-05, | |
| "loss": 1.1494, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.27529249827942187, | |
| "grad_norm": 0.8776836395263672, | |
| "learning_rate": 3.73846699787083e-05, | |
| "loss": 1.1178, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.2924982794218858, | |
| "grad_norm": 0.8656069040298462, | |
| "learning_rate": 3.649751596877218e-05, | |
| "loss": 1.1265, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.3097040605643496, | |
| "grad_norm": 0.9538426995277405, | |
| "learning_rate": 3.561036195883606e-05, | |
| "loss": 1.1138, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.3269098417068135, | |
| "grad_norm": 0.7584338188171387, | |
| "learning_rate": 3.4723207948899934e-05, | |
| "loss": 1.1484, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.3441156228492774, | |
| "grad_norm": 0.9330400228500366, | |
| "learning_rate": 3.383605393896381e-05, | |
| "loss": 1.1758, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3441156228492774, | |
| "eval_loss": 1.2882763147354126, | |
| "eval_runtime": 96.0412, | |
| "eval_samples_per_second": 7.643, | |
| "eval_steps_per_second": 1.531, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.36132140399174123, | |
| "grad_norm": 0.76217120885849, | |
| "learning_rate": 3.2948899929027684e-05, | |
| "loss": 1.1237, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.3785271851342051, | |
| "grad_norm": 0.9417053461074829, | |
| "learning_rate": 3.206174591909156e-05, | |
| "loss": 1.1834, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.395732966276669, | |
| "grad_norm": 0.7688853144645691, | |
| "learning_rate": 3.1174591909155435e-05, | |
| "loss": 1.2116, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.41293874741913283, | |
| "grad_norm": 0.7310340404510498, | |
| "learning_rate": 3.0287437899219306e-05, | |
| "loss": 1.1458, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.4301445285615967, | |
| "grad_norm": 0.8664773106575012, | |
| "learning_rate": 2.940028388928318e-05, | |
| "loss": 1.1651, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.4473503097040606, | |
| "grad_norm": 0.8525294661521912, | |
| "learning_rate": 2.8513129879347057e-05, | |
| "loss": 1.1609, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.46455609084652444, | |
| "grad_norm": 0.9533822536468506, | |
| "learning_rate": 2.7625975869410932e-05, | |
| "loss": 1.1114, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.4817618719889883, | |
| "grad_norm": 0.9051440358161926, | |
| "learning_rate": 2.6738821859474804e-05, | |
| "loss": 1.1417, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.4989676531314522, | |
| "grad_norm": 0.8630899786949158, | |
| "learning_rate": 2.585166784953868e-05, | |
| "loss": 1.1121, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.516173434273916, | |
| "grad_norm": 0.7944740653038025, | |
| "learning_rate": 2.4964513839602558e-05, | |
| "loss": 1.1243, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5333792154163799, | |
| "grad_norm": 0.8027036190032959, | |
| "learning_rate": 2.4077359829666433e-05, | |
| "loss": 1.1151, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.5505849965588437, | |
| "grad_norm": 0.9314497709274292, | |
| "learning_rate": 2.3190205819730308e-05, | |
| "loss": 1.1382, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.5677907777013076, | |
| "grad_norm": 0.8070854544639587, | |
| "learning_rate": 2.2303051809794183e-05, | |
| "loss": 1.1339, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.5849965588437716, | |
| "grad_norm": 0.9217064380645752, | |
| "learning_rate": 2.1415897799858055e-05, | |
| "loss": 1.1411, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.6022023399862354, | |
| "grad_norm": 0.8798492550849915, | |
| "learning_rate": 2.052874378992193e-05, | |
| "loss": 1.1843, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.6194081211286993, | |
| "grad_norm": 0.7490786910057068, | |
| "learning_rate": 1.9641589779985805e-05, | |
| "loss": 1.1571, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.6366139022711631, | |
| "grad_norm": 0.8831539154052734, | |
| "learning_rate": 1.875443577004968e-05, | |
| "loss": 1.1452, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.653819683413627, | |
| "grad_norm": 0.7337464690208435, | |
| "learning_rate": 1.786728176011356e-05, | |
| "loss": 1.1066, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.6710254645560908, | |
| "grad_norm": 0.8665643334388733, | |
| "learning_rate": 1.698012775017743e-05, | |
| "loss": 1.1761, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.6882312456985548, | |
| "grad_norm": 0.9193384051322937, | |
| "learning_rate": 1.6092973740241306e-05, | |
| "loss": 1.1344, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6882312456985548, | |
| "eval_loss": 1.2713440656661987, | |
| "eval_runtime": 91.1794, | |
| "eval_samples_per_second": 8.05, | |
| "eval_steps_per_second": 1.612, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7054370268410186, | |
| "grad_norm": 0.924618661403656, | |
| "learning_rate": 1.5205819730305181e-05, | |
| "loss": 1.1481, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.7226428079834825, | |
| "grad_norm": 0.9339317679405212, | |
| "learning_rate": 1.4318665720369056e-05, | |
| "loss": 1.1114, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.7398485891259463, | |
| "grad_norm": 0.7737475037574768, | |
| "learning_rate": 1.3431511710432932e-05, | |
| "loss": 1.1446, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.7570543702684102, | |
| "grad_norm": 0.8811545372009277, | |
| "learning_rate": 1.2544357700496807e-05, | |
| "loss": 1.1311, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.774260151410874, | |
| "grad_norm": 0.8200384378433228, | |
| "learning_rate": 1.1657203690560682e-05, | |
| "loss": 1.159, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.791465932553338, | |
| "grad_norm": 0.9544340968132019, | |
| "learning_rate": 1.0770049680624557e-05, | |
| "loss": 1.1244, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.8086717136958018, | |
| "grad_norm": 0.8595368266105652, | |
| "learning_rate": 9.88289567068843e-06, | |
| "loss": 1.1819, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.8258774948382657, | |
| "grad_norm": 0.7589201331138611, | |
| "learning_rate": 8.995741660752308e-06, | |
| "loss": 1.0796, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.8430832759807295, | |
| "grad_norm": 0.7756196856498718, | |
| "learning_rate": 8.108587650816183e-06, | |
| "loss": 1.1882, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.8602890571231934, | |
| "grad_norm": 0.9547609686851501, | |
| "learning_rate": 7.221433640880057e-06, | |
| "loss": 1.1521, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8774948382656572, | |
| "grad_norm": 1.0288267135620117, | |
| "learning_rate": 6.3342796309439315e-06, | |
| "loss": 1.1987, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.8947006194081212, | |
| "grad_norm": 1.0321507453918457, | |
| "learning_rate": 5.4471256210078075e-06, | |
| "loss": 1.1331, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.911906400550585, | |
| "grad_norm": 0.8565710783004761, | |
| "learning_rate": 4.559971611071683e-06, | |
| "loss": 1.1583, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.9291121816930489, | |
| "grad_norm": 0.7842207551002502, | |
| "learning_rate": 3.672817601135557e-06, | |
| "loss": 1.1586, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.9463179628355127, | |
| "grad_norm": 0.9376333355903625, | |
| "learning_rate": 2.7856635911994322e-06, | |
| "loss": 1.1702, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.9635237439779766, | |
| "grad_norm": 0.7735033631324768, | |
| "learning_rate": 1.8985095812633074e-06, | |
| "loss": 1.1598, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.9807295251204404, | |
| "grad_norm": 0.8405628204345703, | |
| "learning_rate": 1.0113555713271824e-06, | |
| "loss": 1.1977, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.9979353062629044, | |
| "grad_norm": 0.9018322825431824, | |
| "learning_rate": 1.242015613910575e-07, | |
| "loss": 1.1157, | |
| "step": 2900 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 2906, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3000, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 8, | |
| "early_stopping_threshold": 0.01 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.617784650549304e+18, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |