{ "best_metric": null, "best_model_checkpoint": null, "epoch": 5.0, "eval_steps": 500, "global_step": 25070, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0997207818109294, "grad_norm": 2.346997022628784, "learning_rate": 4.5023932987634625e-05, "loss": 8.2424, "step": 500 }, { "epoch": 0.1994415636218588, "grad_norm": 2.3684158325195312, "learning_rate": 4.0037893897088155e-05, "loss": 7.6851, "step": 1000 }, { "epoch": 0.2991623454327882, "grad_norm": 3.409303665161133, "learning_rate": 3.5051854806541686e-05, "loss": 7.4872, "step": 1500 }, { "epoch": 0.3988831272437176, "grad_norm": 2.615360975265503, "learning_rate": 3.0065815715995216e-05, "loss": 7.344, "step": 2000 }, { "epoch": 0.49860390905464697, "grad_norm": 3.5242176055908203, "learning_rate": 2.5079776625448743e-05, "loss": 7.2749, "step": 2500 }, { "epoch": 0.5983246908655764, "grad_norm": 3.690262794494629, "learning_rate": 2.0093737534902273e-05, "loss": 7.1657, "step": 3000 }, { "epoch": 0.6980454726765057, "grad_norm": 2.940692663192749, "learning_rate": 1.5107698444355806e-05, "loss": 7.1298, "step": 3500 }, { "epoch": 0.7977662544874352, "grad_norm": 2.9132378101348877, "learning_rate": 1.0121659353809334e-05, "loss": 7.0938, "step": 4000 }, { "epoch": 0.8974870362983646, "grad_norm": 3.101921558380127, "learning_rate": 5.135620263262864e-06, "loss": 7.0715, "step": 4500 }, { "epoch": 0.9972078181092939, "grad_norm": 3.2258358001708984, "learning_rate": 1.495811727163941e-07, "loss": 7.0478, "step": 5000 }, { "epoch": 1.0969285999202234, "grad_norm": 3.2722208499908447, "learning_rate": 3.903270841643399e-05, "loss": 7.0374, "step": 5500 }, { "epoch": 1.1966493817311528, "grad_norm": 5.218217849731445, "learning_rate": 3.803550059832469e-05, "loss": 7.0289, "step": 6000 }, { "epoch": 1.2963701635420821, "grad_norm": 3.466571807861328, "learning_rate": 3.70382927802154e-05, "loss": 6.9595, "step": 6500 }, { "epoch": 1.3960909453530115, "grad_norm": 3.688443183898926, "learning_rate": 3.6041084962106106e-05, "loss": 6.9267, "step": 7000 }, { "epoch": 1.4958117271639408, "grad_norm": 3.0426700115203857, "learning_rate": 3.504387714399681e-05, "loss": 6.8954, "step": 7500 }, { "epoch": 1.5955325089748702, "grad_norm": 3.7769949436187744, "learning_rate": 3.404666932588751e-05, "loss": 6.8657, "step": 8000 }, { "epoch": 1.6952532907857998, "grad_norm": 3.0776305198669434, "learning_rate": 3.304946150777822e-05, "loss": 6.8285, "step": 8500 }, { "epoch": 1.7949740725967291, "grad_norm": 3.350515604019165, "learning_rate": 3.2052253689668926e-05, "loss": 6.7948, "step": 9000 }, { "epoch": 1.8946948544076585, "grad_norm": 3.393035411834717, "learning_rate": 3.1055045871559636e-05, "loss": 6.7725, "step": 9500 }, { "epoch": 1.994415636218588, "grad_norm": 3.438401222229004, "learning_rate": 3.0057838053450336e-05, "loss": 6.7484, "step": 10000 }, { "epoch": 2.0941364180295174, "grad_norm": 4.042023181915283, "learning_rate": 2.9060630235341047e-05, "loss": 6.6939, "step": 10500 }, { "epoch": 2.193857199840447, "grad_norm": 3.3481028079986572, "learning_rate": 2.8063422417231757e-05, "loss": 6.6854, "step": 11000 }, { "epoch": 2.293577981651376, "grad_norm": 3.266961097717285, "learning_rate": 2.706820901475868e-05, "loss": 6.6555, "step": 11500 }, { "epoch": 2.3932987634623055, "grad_norm": 3.215405225753784, "learning_rate": 2.607100119664938e-05, "loss": 6.6713, "step": 12000 }, { "epoch": 2.493019545273235, "grad_norm": 3.380500316619873, "learning_rate": 2.507379337854009e-05, "loss": 6.6581, "step": 12500 }, { "epoch": 2.5927403270841642, "grad_norm": 3.536166191101074, "learning_rate": 2.4076585560430796e-05, "loss": 6.5945, "step": 13000 }, { "epoch": 2.6924611088950936, "grad_norm": 3.9319474697113037, "learning_rate": 2.30793777423215e-05, "loss": 6.6057, "step": 13500 }, { "epoch": 2.792181890706023, "grad_norm": 4.334239482879639, "learning_rate": 2.2084164339848425e-05, "loss": 6.5818, "step": 14000 }, { "epoch": 2.8919026725169523, "grad_norm": 4.093286514282227, "learning_rate": 2.1086956521739132e-05, "loss": 6.5732, "step": 14500 }, { "epoch": 2.9916234543278817, "grad_norm": 4.026576995849609, "learning_rate": 2.008974870362984e-05, "loss": 6.5627, "step": 15000 }, { "epoch": 3.0913442361388115, "grad_norm": 3.7285637855529785, "learning_rate": 1.9092540885520542e-05, "loss": 6.5268, "step": 15500 }, { "epoch": 3.191065017949741, "grad_norm": 3.7349226474761963, "learning_rate": 1.809533306741125e-05, "loss": 6.5388, "step": 16000 }, { "epoch": 3.29078579976067, "grad_norm": 3.5330066680908203, "learning_rate": 1.7098125249301956e-05, "loss": 6.5141, "step": 16500 }, { "epoch": 3.3905065815715996, "grad_norm": 3.6961631774902344, "learning_rate": 1.6100917431192662e-05, "loss": 6.5013, "step": 17000 }, { "epoch": 3.490227363382529, "grad_norm": 3.413053274154663, "learning_rate": 1.5103709613083367e-05, "loss": 6.4932, "step": 17500 }, { "epoch": 3.5899481451934583, "grad_norm": 4.584457874298096, "learning_rate": 1.4108496210610292e-05, "loss": 6.4695, "step": 18000 }, { "epoch": 3.6896689270043876, "grad_norm": 3.3078787326812744, "learning_rate": 1.3111288392500998e-05, "loss": 6.4711, "step": 18500 }, { "epoch": 3.789389708815317, "grad_norm": 3.6679279804229736, "learning_rate": 1.2114080574391703e-05, "loss": 6.466, "step": 19000 }, { "epoch": 3.8891104906262464, "grad_norm": 4.358784198760986, "learning_rate": 1.1116872756282408e-05, "loss": 6.4568, "step": 19500 }, { "epoch": 3.988831272437176, "grad_norm": 4.014244556427002, "learning_rate": 1.0119664938173115e-05, "loss": 6.4536, "step": 20000 }, { "epoch": 4.0885520542481055, "grad_norm": 3.8396079540252686, "learning_rate": 9.122457120063822e-06, "loss": 6.443, "step": 20500 }, { "epoch": 4.188272836059035, "grad_norm": 3.850647449493408, "learning_rate": 8.125249301954529e-06, "loss": 6.4186, "step": 21000 }, { "epoch": 4.287993617869964, "grad_norm": 3.829951047897339, "learning_rate": 7.128041483845234e-06, "loss": 6.4178, "step": 21500 }, { "epoch": 4.387714399680894, "grad_norm": 3.5512278079986572, "learning_rate": 6.132828081372159e-06, "loss": 6.4055, "step": 22000 }, { "epoch": 4.487435181491823, "grad_norm": 3.568665027618408, "learning_rate": 5.135620263262864e-06, "loss": 6.4076, "step": 22500 }, { "epoch": 4.587155963302752, "grad_norm": 3.71463942527771, "learning_rate": 4.13841244515357e-06, "loss": 6.4086, "step": 23000 }, { "epoch": 4.686876745113682, "grad_norm": 3.9615983963012695, "learning_rate": 3.1412046270442757e-06, "loss": 6.4061, "step": 23500 }, { "epoch": 4.786597526924611, "grad_norm": 4.0287909507751465, "learning_rate": 2.1459912245712007e-06, "loss": 6.3772, "step": 24000 }, { "epoch": 4.88631830873554, "grad_norm": 4.012565612792969, "learning_rate": 1.1487834064619066e-06, "loss": 6.3956, "step": 24500 }, { "epoch": 4.98603909054647, "grad_norm": 4.36814022064209, "learning_rate": 1.515755883526127e-07, "loss": 6.3996, "step": 25000 }, { "epoch": 5.0, "step": 25070, "total_flos": 2.639861525017728e+16, "train_loss": 5.285465436767286, "train_runtime": 6500.188, "train_samples_per_second": 61.705, "train_steps_per_second": 3.857 } ], "logging_steps": 500, "max_steps": 25070, "num_input_tokens_seen": 0, "num_train_epochs": 5, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 2.639861525017728e+16, "train_batch_size": 16, "trial_name": null, "trial_params": null }