{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.0, "eval_steps": 500, "global_step": 76, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.013157894736842105, "grad_norm": 1700.623831965343, "learning_rate": 1.25e-06, "loss": 11.6905, "step": 1 }, { "epoch": 0.02631578947368421, "grad_norm": 1649.0192864629867, "learning_rate": 2.5e-06, "loss": 11.7067, "step": 2 }, { "epoch": 0.039473684210526314, "grad_norm": 1101.5446707599804, "learning_rate": 3.7500000000000005e-06, "loss": 10.8183, "step": 3 }, { "epoch": 0.05263157894736842, "grad_norm": 1102.519048385476, "learning_rate": 5e-06, "loss": 9.0508, "step": 4 }, { "epoch": 0.06578947368421052, "grad_norm": 802.3239878863764, "learning_rate": 6.25e-06, "loss": 5.9038, "step": 5 }, { "epoch": 0.07894736842105263, "grad_norm": 258.11430612786404, "learning_rate": 7.500000000000001e-06, "loss": 4.7947, "step": 6 }, { "epoch": 0.09210526315789473, "grad_norm": 403.267159818518, "learning_rate": 8.750000000000001e-06, "loss": 5.461, "step": 7 }, { "epoch": 0.10526315789473684, "grad_norm": 98.17603349248624, "learning_rate": 1e-05, "loss": 3.1841, "step": 8 }, { "epoch": 0.11842105263157894, "grad_norm": 125.08383294967409, "learning_rate": 9.994664874011864e-06, "loss": 5.2682, "step": 9 }, { "epoch": 0.13157894736842105, "grad_norm": 178.4978710757842, "learning_rate": 9.978670881475173e-06, "loss": 4.6418, "step": 10 }, { "epoch": 0.14473684210526316, "grad_norm": 33.98589110631106, "learning_rate": 9.952052154376027e-06, "loss": 3.3566, "step": 11 }, { "epoch": 0.15789473684210525, "grad_norm": 40.35276750130872, "learning_rate": 9.91486549841951e-06, "loss": 3.6266, "step": 12 }, { "epoch": 0.17105263157894737, "grad_norm": 26.024781858769245, "learning_rate": 9.867190271803466e-06, "loss": 2.921, "step": 13 }, { "epoch": 0.18421052631578946, "grad_norm": 39.19155525491511, "learning_rate": 9.809128215864096e-06, "loss": 3.3999, "step": 14 }, { "epoch": 0.19736842105263158, "grad_norm": 31.392187014966936, "learning_rate": 9.74080323795483e-06, "loss": 2.9989, "step": 15 }, { "epoch": 0.21052631578947367, "grad_norm": 11.487293577006643, "learning_rate": 9.66236114702178e-06, "loss": 2.4908, "step": 16 }, { "epoch": 0.2236842105263158, "grad_norm": 23.906559218300867, "learning_rate": 9.573969342440107e-06, "loss": 2.3537, "step": 17 }, { "epoch": 0.23684210526315788, "grad_norm": 18.145322592938253, "learning_rate": 9.475816456775313e-06, "loss": 2.3298, "step": 18 }, { "epoch": 0.25, "grad_norm": 19.690523946140633, "learning_rate": 9.368111953231849e-06, "loss": 2.1922, "step": 19 }, { "epoch": 0.2631578947368421, "grad_norm": 9.140283229170583, "learning_rate": 9.251085678648072e-06, "loss": 2.0839, "step": 20 }, { "epoch": 0.27631578947368424, "grad_norm": 17.542896607710595, "learning_rate": 9.124987372991512e-06, "loss": 2.1548, "step": 21 }, { "epoch": 0.2894736842105263, "grad_norm": 14.27679773227752, "learning_rate": 8.990086136401199e-06, "loss": 2.0888, "step": 22 }, { "epoch": 0.3026315789473684, "grad_norm": 16.817965246454907, "learning_rate": 8.846669854914395e-06, "loss": 1.8762, "step": 23 }, { "epoch": 0.3157894736842105, "grad_norm": 10.564762547828062, "learning_rate": 8.695044586103297e-06, "loss": 1.8733, "step": 24 }, { "epoch": 0.32894736842105265, "grad_norm": 14.138309552512082, "learning_rate": 8.535533905932739e-06, "loss": 1.7913, "step": 25 }, { "epoch": 0.34210526315789475, "grad_norm": 20.034746365001553, "learning_rate": 8.368478218232787e-06, "loss": 1.737, "step": 26 }, { "epoch": 0.35526315789473684, "grad_norm": 16.855851572810284, "learning_rate": 8.194234028259806e-06, "loss": 1.6829, "step": 27 }, { "epoch": 0.3684210526315789, "grad_norm": 16.950076744646207, "learning_rate": 8.013173181896283e-06, "loss": 1.4163, "step": 28 }, { "epoch": 0.3815789473684211, "grad_norm": 22.124680222566646, "learning_rate": 7.82568207211296e-06, "loss": 1.5623, "step": 29 }, { "epoch": 0.39473684210526316, "grad_norm": 28.487415904850273, "learning_rate": 7.63216081438678e-06, "loss": 1.4885, "step": 30 }, { "epoch": 0.40789473684210525, "grad_norm": 43.5618784839132, "learning_rate": 7.4330223928342814e-06, "loss": 1.5518, "step": 31 }, { "epoch": 0.42105263157894735, "grad_norm": 24.171098856189804, "learning_rate": 7.2286917788826926e-06, "loss": 1.4394, "step": 32 }, { "epoch": 0.4342105263157895, "grad_norm": 23.119722073349426, "learning_rate": 7.019605024359475e-06, "loss": 1.2995, "step": 33 }, { "epoch": 0.4473684210526316, "grad_norm": 16.689485734152086, "learning_rate": 6.806208330935766e-06, "loss": 1.1763, "step": 34 }, { "epoch": 0.4605263157894737, "grad_norm": 15.859289728493408, "learning_rate": 6.588957097909509e-06, "loss": 1.144, "step": 35 }, { "epoch": 0.47368421052631576, "grad_norm": 15.73886933600492, "learning_rate": 6.368314950360416e-06, "loss": 1.1619, "step": 36 }, { "epoch": 0.4868421052631579, "grad_norm": 15.1571280026014, "learning_rate": 6.144752749750671e-06, "loss": 1.0688, "step": 37 }, { "epoch": 0.5, "grad_norm": 13.225349267137519, "learning_rate": 5.918747589082853e-06, "loss": 1.1828, "step": 38 }, { "epoch": 0.5131578947368421, "grad_norm": 14.492320032343926, "learning_rate": 5.690781774759412e-06, "loss": 1.0862, "step": 39 }, { "epoch": 0.5263157894736842, "grad_norm": 9.45491415828414, "learning_rate": 5.46134179731651e-06, "loss": 1.0357, "step": 40 }, { "epoch": 0.5394736842105263, "grad_norm": 11.676684639895935, "learning_rate": 5.230917293228699e-06, "loss": 0.952, "step": 41 }, { "epoch": 0.5526315789473685, "grad_norm": 17.496877680196757, "learning_rate": 5e-06, "loss": 0.9923, "step": 42 }, { "epoch": 0.5657894736842105, "grad_norm": 21.183543111805434, "learning_rate": 4.7690827067713035e-06, "loss": 1.1084, "step": 43 }, { "epoch": 0.5789473684210527, "grad_norm": 16.65044124544019, "learning_rate": 4.53865820268349e-06, "loss": 1.0066, "step": 44 }, { "epoch": 0.5921052631578947, "grad_norm": 14.428813163253018, "learning_rate": 4.309218225240591e-06, "loss": 0.8975, "step": 45 }, { "epoch": 0.6052631578947368, "grad_norm": 13.962879418934243, "learning_rate": 4.081252410917148e-06, "loss": 0.9082, "step": 46 }, { "epoch": 0.618421052631579, "grad_norm": 14.852973745352799, "learning_rate": 3.855247250249331e-06, "loss": 0.7634, "step": 47 }, { "epoch": 0.631578947368421, "grad_norm": 16.12742922776505, "learning_rate": 3.6316850496395863e-06, "loss": 0.9353, "step": 48 }, { "epoch": 0.6447368421052632, "grad_norm": 10.839276159848792, "learning_rate": 3.4110429020904924e-06, "loss": 0.8348, "step": 49 }, { "epoch": 0.6578947368421053, "grad_norm": 28.986407002948578, "learning_rate": 3.1937916690642356e-06, "loss": 0.9786, "step": 50 }, { "epoch": 0.6710526315789473, "grad_norm": 9.721255219536264, "learning_rate": 2.980394975640526e-06, "loss": 0.7851, "step": 51 }, { "epoch": 0.6842105263157895, "grad_norm": 14.650668800173477, "learning_rate": 2.771308221117309e-06, "loss": 0.9191, "step": 52 }, { "epoch": 0.6973684210526315, "grad_norm": 12.383256838083554, "learning_rate": 2.5669776071657194e-06, "loss": 0.911, "step": 53 }, { "epoch": 0.7105263157894737, "grad_norm": 9.794724839588078, "learning_rate": 2.3678391856132203e-06, "loss": 0.8739, "step": 54 }, { "epoch": 0.7236842105263158, "grad_norm": 16.630173515158628, "learning_rate": 2.174317927887041e-06, "loss": 0.8075, "step": 55 }, { "epoch": 0.7368421052631579, "grad_norm": 13.198903695310259, "learning_rate": 1.9868268181037186e-06, "loss": 0.9054, "step": 56 }, { "epoch": 0.75, "grad_norm": 12.519325479242989, "learning_rate": 1.8057659717401948e-06, "loss": 0.7372, "step": 57 }, { "epoch": 0.7631578947368421, "grad_norm": 10.337188449964808, "learning_rate": 1.6315217817672142e-06, "loss": 0.7301, "step": 58 }, { "epoch": 0.7763157894736842, "grad_norm": 8.532097408416009, "learning_rate": 1.4644660940672628e-06, "loss": 0.7052, "step": 59 }, { "epoch": 0.7894736842105263, "grad_norm": 9.94979317440902, "learning_rate": 1.3049554138967052e-06, "loss": 0.701, "step": 60 }, { "epoch": 0.8026315789473685, "grad_norm": 14.270660743327786, "learning_rate": 1.1533301450856054e-06, "loss": 0.7537, "step": 61 }, { "epoch": 0.8157894736842105, "grad_norm": 8.630752726293801, "learning_rate": 1.0099138635988026e-06, "loss": 0.7162, "step": 62 }, { "epoch": 0.8289473684210527, "grad_norm": 10.632504441908269, "learning_rate": 8.750126270084891e-07, "loss": 0.7647, "step": 63 }, { "epoch": 0.8421052631578947, "grad_norm": 8.980630672317927, "learning_rate": 7.489143213519301e-07, "loss": 0.6812, "step": 64 }, { "epoch": 0.8552631578947368, "grad_norm": 5.941122650864663, "learning_rate": 6.318880467681527e-07, "loss": 0.6393, "step": 65 }, { "epoch": 0.868421052631579, "grad_norm": 13.118946568151987, "learning_rate": 5.241835432246888e-07, "loss": 0.6937, "step": 66 }, { "epoch": 0.881578947368421, "grad_norm": 11.16288012285256, "learning_rate": 4.2603065755989493e-07, "loss": 0.7598, "step": 67 }, { "epoch": 0.8947368421052632, "grad_norm": 8.752876273254847, "learning_rate": 3.3763885297822153e-07, "loss": 0.6724, "step": 68 }, { "epoch": 0.9078947368421053, "grad_norm": 10.41179108138972, "learning_rate": 2.5919676204517073e-07, "loss": 0.7259, "step": 69 }, { "epoch": 0.9210526315789473, "grad_norm": 8.452683523386511, "learning_rate": 1.908717841359048e-07, "loss": 0.7196, "step": 70 }, { "epoch": 0.9342105263157895, "grad_norm": 9.08322472471614, "learning_rate": 1.328097281965357e-07, "loss": 0.6541, "step": 71 }, { "epoch": 0.9473684210526315, "grad_norm": 11.964823295684173, "learning_rate": 8.513450158049109e-08, "loss": 0.6954, "step": 72 }, { "epoch": 0.9605263157894737, "grad_norm": 6.649921077114204, "learning_rate": 4.794784562397459e-08, "loss": 0.633, "step": 73 }, { "epoch": 0.9736842105263158, "grad_norm": 9.29104017000911, "learning_rate": 2.1329118524827662e-08, "loss": 0.701, "step": 74 }, { "epoch": 0.9868421052631579, "grad_norm": 7.6775727377006655, "learning_rate": 5.3351259881379016e-09, "loss": 0.6241, "step": 75 }, { "epoch": 1.0, "grad_norm": 8.200287140678359, "learning_rate": 0.0, "loss": 0.8084, "step": 76 }, { "epoch": 1.0, "eval_loss": 0.7041537165641785, "eval_runtime": 129.4756, "eval_samples_per_second": 39.428, "eval_steps_per_second": 1.236, "step": 76 }, { "epoch": 1.0, "step": 76, "total_flos": 2.244031216091136e+16, "train_loss": 2.0892962131061052, "train_runtime": 1016.0826, "train_samples_per_second": 9.544, "train_steps_per_second": 0.075 } ], "logging_steps": 1, "max_steps": 76, "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": 2.244031216091136e+16, "train_batch_size": 8, "trial_name": null, "trial_params": null }