| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.25, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "Batch Mean": 2.61431884765625, | |
| "accuracy": 0.4765625, | |
| "epoch": 0, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0.0025, | |
| "grad_norm": 42.59496307373047, | |
| "learning_rate": 1.5000000000000002e-07, | |
| "loss": 0.8092, | |
| "step": 1 | |
| }, | |
| { | |
| "Batch Mean": 2.574005126953125, | |
| "accuracy": 0.578125, | |
| "epoch": 0.0025, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005, | |
| "grad_norm": 46.709930419921875, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 0.7799, | |
| "step": 2 | |
| }, | |
| { | |
| "Batch Mean": 2.560516357421875, | |
| "accuracy": 0.484375, | |
| "epoch": 0.005, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0075, | |
| "grad_norm": 44.51314163208008, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.8043, | |
| "step": 3 | |
| }, | |
| { | |
| "Batch Mean": 2.6197509765625, | |
| "accuracy": 0.5, | |
| "epoch": 0.0075, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 42.86121368408203, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.8014, | |
| "step": 4 | |
| }, | |
| { | |
| "Batch Mean": 2.565338134765625, | |
| "accuracy": 0.53125, | |
| "epoch": 0.01, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0125, | |
| "grad_norm": 44.409908294677734, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.7983, | |
| "step": 5 | |
| }, | |
| { | |
| "Batch Mean": 2.522857666015625, | |
| "accuracy": 0.484375, | |
| "epoch": 0.0125, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015, | |
| "grad_norm": 43.77534484863281, | |
| "learning_rate": 9e-07, | |
| "loss": 0.7975, | |
| "step": 6 | |
| }, | |
| { | |
| "Batch Mean": 2.3875732421875, | |
| "accuracy": 0.4921875, | |
| "epoch": 0.015, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0175, | |
| "grad_norm": 41.981407165527344, | |
| "learning_rate": 1.05e-06, | |
| "loss": 0.7951, | |
| "step": 7 | |
| }, | |
| { | |
| "Batch Mean": 2.317047119140625, | |
| "accuracy": 0.515625, | |
| "epoch": 0.0175, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 40.53751754760742, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.791, | |
| "step": 8 | |
| }, | |
| { | |
| "Batch Mean": 1.757080078125, | |
| "accuracy": 0.5078125, | |
| "epoch": 0.02, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0225, | |
| "grad_norm": 31.105939865112305, | |
| "learning_rate": 1.35e-06, | |
| "loss": 0.7571, | |
| "step": 9 | |
| }, | |
| { | |
| "Batch Mean": 1.6220855712890625, | |
| "accuracy": 0.578125, | |
| "epoch": 0.0225, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 33.61161804199219, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.7234, | |
| "step": 10 | |
| }, | |
| { | |
| "Batch Mean": 1.400146484375, | |
| "accuracy": 0.53125, | |
| "epoch": 0.025, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0275, | |
| "grad_norm": 34.554622650146484, | |
| "learning_rate": 1.65e-06, | |
| "loss": 0.725, | |
| "step": 11 | |
| }, | |
| { | |
| "Batch Mean": -0.43697381019592285, | |
| "accuracy": 0.5234375, | |
| "epoch": 0.0275, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 12.222326278686523, | |
| "learning_rate": 1.8e-06, | |
| "loss": 0.6986, | |
| "step": 12 | |
| }, | |
| { | |
| "Batch Mean": -0.8854951858520508, | |
| "accuracy": 0.5234375, | |
| "epoch": 0.03, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0325, | |
| "grad_norm": 16.398786544799805, | |
| "learning_rate": 1.95e-06, | |
| "loss": 0.7081, | |
| "step": 13 | |
| }, | |
| { | |
| "Batch Mean": -0.9508838653564453, | |
| "accuracy": 0.640625, | |
| "epoch": 0.0325, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.035, | |
| "grad_norm": 19.590002059936523, | |
| "learning_rate": 2.1e-06, | |
| "loss": 0.6784, | |
| "step": 14 | |
| }, | |
| { | |
| "Batch Mean": -1.0560526847839355, | |
| "accuracy": 0.4921875, | |
| "epoch": 0.035, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0375, | |
| "grad_norm": 22.669742584228516, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.7235, | |
| "step": 15 | |
| }, | |
| { | |
| "Batch Mean": -1.1405725479125977, | |
| "accuracy": 0.625, | |
| "epoch": 0.0375, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 22.189626693725586, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.6714, | |
| "step": 16 | |
| }, | |
| { | |
| "Batch Mean": -1.1522831916809082, | |
| "accuracy": 0.640625, | |
| "epoch": 0.04, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0425, | |
| "grad_norm": 19.9060115814209, | |
| "learning_rate": 2.55e-06, | |
| "loss": 0.6757, | |
| "step": 17 | |
| }, | |
| { | |
| "Batch Mean": -1.1666946411132812, | |
| "accuracy": 0.53125, | |
| "epoch": 0.0425, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.045, | |
| "grad_norm": 23.661216735839844, | |
| "learning_rate": 2.7e-06, | |
| "loss": 0.7002, | |
| "step": 18 | |
| }, | |
| { | |
| "Batch Mean": -0.999359130859375, | |
| "accuracy": 0.6640625, | |
| "epoch": 0.045, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0475, | |
| "grad_norm": 15.433859825134277, | |
| "learning_rate": 2.85e-06, | |
| "loss": 0.6471, | |
| "step": 19 | |
| }, | |
| { | |
| "Batch Mean": -0.6625549793243408, | |
| "accuracy": 0.6171875, | |
| "epoch": 0.0475, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 13.135390281677246, | |
| "learning_rate": 3e-06, | |
| "loss": 0.662, | |
| "step": 20 | |
| }, | |
| { | |
| "Batch Mean": -0.13487493991851807, | |
| "accuracy": 0.6171875, | |
| "epoch": 0.05, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0525, | |
| "grad_norm": 7.336737632751465, | |
| "learning_rate": 2.992105263157895e-06, | |
| "loss": 0.629, | |
| "step": 21 | |
| }, | |
| { | |
| "Batch Mean": 0.2938040494918823, | |
| "accuracy": 0.6640625, | |
| "epoch": 0.0525, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.055, | |
| "grad_norm": 9.586155891418457, | |
| "learning_rate": 2.9842105263157896e-06, | |
| "loss": 0.6225, | |
| "step": 22 | |
| }, | |
| { | |
| "Batch Mean": 0.6780391931533813, | |
| "accuracy": 0.7421875, | |
| "epoch": 0.055, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0575, | |
| "grad_norm": 14.825817108154297, | |
| "learning_rate": 2.9763157894736843e-06, | |
| "loss": 0.607, | |
| "step": 23 | |
| }, | |
| { | |
| "Batch Mean": 0.9421095848083496, | |
| "accuracy": 0.609375, | |
| "epoch": 0.0575, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 16.7529354095459, | |
| "learning_rate": 2.968421052631579e-06, | |
| "loss": 0.6542, | |
| "step": 24 | |
| }, | |
| { | |
| "Batch Mean": 1.0722179412841797, | |
| "accuracy": 0.71875, | |
| "epoch": 0.06, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 18.940032958984375, | |
| "learning_rate": 2.960526315789474e-06, | |
| "loss": 0.5882, | |
| "step": 25 | |
| }, | |
| { | |
| "Batch Mean": 0.8396664261817932, | |
| "accuracy": 0.6484375, | |
| "epoch": 0.0625, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.065, | |
| "grad_norm": 14.680763244628906, | |
| "learning_rate": 2.9526315789473685e-06, | |
| "loss": 0.6463, | |
| "step": 26 | |
| }, | |
| { | |
| "Batch Mean": 0.2986793518066406, | |
| "accuracy": 0.671875, | |
| "epoch": 0.065, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0675, | |
| "grad_norm": 7.3362345695495605, | |
| "learning_rate": 2.9447368421052633e-06, | |
| "loss": 0.5923, | |
| "step": 27 | |
| }, | |
| { | |
| "Batch Mean": -0.29196763038635254, | |
| "accuracy": 0.734375, | |
| "epoch": 0.0675, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 8.053507804870605, | |
| "learning_rate": 2.936842105263158e-06, | |
| "loss": 0.5261, | |
| "step": 28 | |
| }, | |
| { | |
| "Batch Mean": -0.8161113262176514, | |
| "accuracy": 0.7109375, | |
| "epoch": 0.07, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0725, | |
| "grad_norm": 14.036270141601562, | |
| "learning_rate": 2.9289473684210528e-06, | |
| "loss": 0.5482, | |
| "step": 29 | |
| }, | |
| { | |
| "Batch Mean": -0.7824737429618835, | |
| "accuracy": 0.75, | |
| "epoch": 0.0725, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 13.984892845153809, | |
| "learning_rate": 2.9210526315789475e-06, | |
| "loss": 0.5939, | |
| "step": 30 | |
| }, | |
| { | |
| "Batch Mean": -0.4927825927734375, | |
| "accuracy": 0.734375, | |
| "epoch": 0.075, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0775, | |
| "grad_norm": 10.751565933227539, | |
| "learning_rate": 2.9131578947368423e-06, | |
| "loss": 0.5901, | |
| "step": 31 | |
| }, | |
| { | |
| "Batch Mean": 0.15497040748596191, | |
| "accuracy": 0.6796875, | |
| "epoch": 0.0775, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 8.30102252960205, | |
| "learning_rate": 2.905263157894737e-06, | |
| "loss": 0.5731, | |
| "step": 32 | |
| }, | |
| { | |
| "Batch Mean": 0.16421844065189362, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.08, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0825, | |
| "grad_norm": 9.087366104125977, | |
| "learning_rate": 2.8973684210526318e-06, | |
| "loss": 0.4958, | |
| "step": 33 | |
| }, | |
| { | |
| "Batch Mean": 0.3596491813659668, | |
| "accuracy": 0.859375, | |
| "epoch": 0.0825, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.085, | |
| "grad_norm": 9.984371185302734, | |
| "learning_rate": 2.8894736842105265e-06, | |
| "loss": 0.4321, | |
| "step": 34 | |
| }, | |
| { | |
| "Batch Mean": 0.2128266543149948, | |
| "accuracy": 0.734375, | |
| "epoch": 0.085, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0875, | |
| "grad_norm": 7.400077819824219, | |
| "learning_rate": 2.8815789473684213e-06, | |
| "loss": 0.5178, | |
| "step": 35 | |
| }, | |
| { | |
| "Batch Mean": 0.10449030995368958, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.0875, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 6.669457912445068, | |
| "learning_rate": 2.873684210526316e-06, | |
| "loss": 0.4914, | |
| "step": 36 | |
| }, | |
| { | |
| "Batch Mean": -0.24382781982421875, | |
| "accuracy": 0.6953125, | |
| "epoch": 0.09, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0925, | |
| "grad_norm": 10.831747055053711, | |
| "learning_rate": 2.8657894736842103e-06, | |
| "loss": 0.5305, | |
| "step": 37 | |
| }, | |
| { | |
| "Batch Mean": -0.29154396057128906, | |
| "accuracy": 0.7109375, | |
| "epoch": 0.0925, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.095, | |
| "grad_norm": 11.060967445373535, | |
| "learning_rate": 2.857894736842105e-06, | |
| "loss": 0.5589, | |
| "step": 38 | |
| }, | |
| { | |
| "Batch Mean": -0.07830595970153809, | |
| "accuracy": 0.765625, | |
| "epoch": 0.095, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0975, | |
| "grad_norm": 8.426444053649902, | |
| "learning_rate": 2.85e-06, | |
| "loss": 0.4663, | |
| "step": 39 | |
| }, | |
| { | |
| "Batch Mean": -0.09856069087982178, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.0975, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 11.046717643737793, | |
| "learning_rate": 2.8421052631578946e-06, | |
| "loss": 0.455, | |
| "step": 40 | |
| }, | |
| { | |
| "Batch Mean": 0.29904642701148987, | |
| "accuracy": 0.8125, | |
| "epoch": 0.1, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1025, | |
| "grad_norm": 11.232060432434082, | |
| "learning_rate": 2.8342105263157897e-06, | |
| "loss": 0.437, | |
| "step": 41 | |
| }, | |
| { | |
| "Batch Mean": 0.43527090549468994, | |
| "accuracy": 0.78125, | |
| "epoch": 0.1025, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.105, | |
| "grad_norm": 12.555906295776367, | |
| "learning_rate": 2.8263157894736845e-06, | |
| "loss": 0.5316, | |
| "step": 42 | |
| }, | |
| { | |
| "Batch Mean": 0.02639901638031006, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.105, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1075, | |
| "grad_norm": 9.500948905944824, | |
| "learning_rate": 2.8184210526315792e-06, | |
| "loss": 0.4874, | |
| "step": 43 | |
| }, | |
| { | |
| "Batch Mean": -0.3524761199951172, | |
| "accuracy": 0.7421875, | |
| "epoch": 0.1075, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 10.417457580566406, | |
| "learning_rate": 2.810526315789474e-06, | |
| "loss": 0.5011, | |
| "step": 44 | |
| }, | |
| { | |
| "Batch Mean": -0.32391130924224854, | |
| "accuracy": 0.8125, | |
| "epoch": 0.11, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1125, | |
| "grad_norm": 9.867128372192383, | |
| "learning_rate": 2.8026315789473687e-06, | |
| "loss": 0.4908, | |
| "step": 45 | |
| }, | |
| { | |
| "Batch Mean": -0.42060422897338867, | |
| "accuracy": 0.8125, | |
| "epoch": 0.1125, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.115, | |
| "grad_norm": 10.613035202026367, | |
| "learning_rate": 2.7947368421052635e-06, | |
| "loss": 0.4642, | |
| "step": 46 | |
| }, | |
| { | |
| "Batch Mean": 0.04145359992980957, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.115, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.1175, | |
| "grad_norm": 12.13752555847168, | |
| "learning_rate": 2.7868421052631578e-06, | |
| "loss": 0.4943, | |
| "step": 47 | |
| }, | |
| { | |
| "Batch Mean": 0.24553179740905762, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1175, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 11.048128128051758, | |
| "learning_rate": 2.7789473684210525e-06, | |
| "loss": 0.3946, | |
| "step": 48 | |
| }, | |
| { | |
| "Batch Mean": -0.05052506923675537, | |
| "accuracy": 0.796875, | |
| "epoch": 0.12, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1225, | |
| "grad_norm": 9.053565979003906, | |
| "learning_rate": 2.7710526315789473e-06, | |
| "loss": 0.4474, | |
| "step": 49 | |
| }, | |
| { | |
| "Batch Mean": -0.43336963653564453, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.1225, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 15.04668140411377, | |
| "learning_rate": 2.763157894736842e-06, | |
| "loss": 0.5496, | |
| "step": 50 | |
| }, | |
| { | |
| "Batch Mean": -0.29952335357666016, | |
| "accuracy": 0.765625, | |
| "epoch": 0.125, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1275, | |
| "grad_norm": 11.302435874938965, | |
| "learning_rate": 2.7552631578947368e-06, | |
| "loss": 0.4337, | |
| "step": 51 | |
| }, | |
| { | |
| "Batch Mean": -0.01873302459716797, | |
| "accuracy": 0.828125, | |
| "epoch": 0.1275, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 10.582301139831543, | |
| "learning_rate": 2.7473684210526315e-06, | |
| "loss": 0.4073, | |
| "step": 52 | |
| }, | |
| { | |
| "Batch Mean": 0.4117751717567444, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.13, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.1325, | |
| "grad_norm": 9.776220321655273, | |
| "learning_rate": 2.7394736842105263e-06, | |
| "loss": 0.4235, | |
| "step": 53 | |
| }, | |
| { | |
| "Batch Mean": 0.5990171432495117, | |
| "accuracy": 0.78125, | |
| "epoch": 0.1325, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.135, | |
| "grad_norm": 12.647154808044434, | |
| "learning_rate": 2.7315789473684214e-06, | |
| "loss": 0.4858, | |
| "step": 54 | |
| }, | |
| { | |
| "Batch Mean": 0.2538492679595947, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.135, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1375, | |
| "grad_norm": 11.564103126525879, | |
| "learning_rate": 2.723684210526316e-06, | |
| "loss": 0.5233, | |
| "step": 55 | |
| }, | |
| { | |
| "Batch Mean": 0.17059040069580078, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.1375, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 10.589515686035156, | |
| "learning_rate": 2.715789473684211e-06, | |
| "loss": 0.4762, | |
| "step": 56 | |
| }, | |
| { | |
| "Batch Mean": -0.3642357587814331, | |
| "accuracy": 0.78125, | |
| "epoch": 0.14, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1425, | |
| "grad_norm": 10.727502822875977, | |
| "learning_rate": 2.7078947368421052e-06, | |
| "loss": 0.5289, | |
| "step": 57 | |
| }, | |
| { | |
| "Batch Mean": -0.5980481505393982, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.1425, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.145, | |
| "grad_norm": 8.915611267089844, | |
| "learning_rate": 2.7e-06, | |
| "loss": 0.4038, | |
| "step": 58 | |
| }, | |
| { | |
| "Batch Mean": -0.46013855934143066, | |
| "accuracy": 0.859375, | |
| "epoch": 0.145, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.1475, | |
| "grad_norm": 9.178686141967773, | |
| "learning_rate": 2.6921052631578947e-06, | |
| "loss": 0.3708, | |
| "step": 59 | |
| }, | |
| { | |
| "Batch Mean": -0.1478586494922638, | |
| "accuracy": 0.84375, | |
| "epoch": 0.1475, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 6.339688301086426, | |
| "learning_rate": 2.6842105263157895e-06, | |
| "loss": 0.4036, | |
| "step": 60 | |
| }, | |
| { | |
| "Batch Mean": 0.003389716148376465, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.15, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1525, | |
| "grad_norm": 6.763300895690918, | |
| "learning_rate": 2.6763157894736842e-06, | |
| "loss": 0.5163, | |
| "step": 61 | |
| }, | |
| { | |
| "Batch Mean": 0.4018087387084961, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.1525, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.155, | |
| "grad_norm": 8.631525039672852, | |
| "learning_rate": 2.668421052631579e-06, | |
| "loss": 0.4219, | |
| "step": 62 | |
| }, | |
| { | |
| "Batch Mean": 0.342004656791687, | |
| "accuracy": 0.734375, | |
| "epoch": 0.155, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.1575, | |
| "grad_norm": 9.190362930297852, | |
| "learning_rate": 2.6605263157894737e-06, | |
| "loss": 0.4708, | |
| "step": 63 | |
| }, | |
| { | |
| "Batch Mean": 0.5178697109222412, | |
| "accuracy": 0.8125, | |
| "epoch": 0.1575, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 10.207452774047852, | |
| "learning_rate": 2.6526315789473685e-06, | |
| "loss": 0.4458, | |
| "step": 64 | |
| }, | |
| { | |
| "Batch Mean": 0.03012150526046753, | |
| "accuracy": 0.8125, | |
| "epoch": 0.16, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1625, | |
| "grad_norm": 8.050507545471191, | |
| "learning_rate": 2.644736842105263e-06, | |
| "loss": 0.4344, | |
| "step": 65 | |
| }, | |
| { | |
| "Batch Mean": -0.3691895604133606, | |
| "accuracy": 0.78125, | |
| "epoch": 0.1625, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.165, | |
| "grad_norm": 11.163555145263672, | |
| "learning_rate": 2.636842105263158e-06, | |
| "loss": 0.4276, | |
| "step": 66 | |
| }, | |
| { | |
| "Batch Mean": -0.2624788284301758, | |
| "accuracy": 0.8125, | |
| "epoch": 0.165, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.1675, | |
| "grad_norm": 9.523059844970703, | |
| "learning_rate": 2.6289473684210527e-06, | |
| "loss": 0.4001, | |
| "step": 67 | |
| }, | |
| { | |
| "Batch Mean": -0.0055138468742370605, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1675, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 9.094910621643066, | |
| "learning_rate": 2.6210526315789474e-06, | |
| "loss": 0.4024, | |
| "step": 68 | |
| }, | |
| { | |
| "Batch Mean": 0.2572704553604126, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.17, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.1725, | |
| "grad_norm": 10.00645923614502, | |
| "learning_rate": 2.613157894736842e-06, | |
| "loss": 0.4802, | |
| "step": 69 | |
| }, | |
| { | |
| "Batch Mean": 0.5112218856811523, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1725, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 11.776649475097656, | |
| "learning_rate": 2.605263157894737e-06, | |
| "loss": 0.4194, | |
| "step": 70 | |
| }, | |
| { | |
| "Batch Mean": 0.29661768674850464, | |
| "accuracy": 0.8125, | |
| "epoch": 0.175, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1775, | |
| "grad_norm": 9.701162338256836, | |
| "learning_rate": 2.5973684210526317e-06, | |
| "loss": 0.4356, | |
| "step": 71 | |
| }, | |
| { | |
| "Batch Mean": 0.049837589263916016, | |
| "accuracy": 0.859375, | |
| "epoch": 0.1775, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 8.692646980285645, | |
| "learning_rate": 2.5894736842105264e-06, | |
| "loss": 0.2881, | |
| "step": 72 | |
| }, | |
| { | |
| "Batch Mean": -0.3786022663116455, | |
| "accuracy": 0.8125, | |
| "epoch": 0.18, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1825, | |
| "grad_norm": 10.834145545959473, | |
| "learning_rate": 2.581578947368421e-06, | |
| "loss": 0.4654, | |
| "step": 73 | |
| }, | |
| { | |
| "Batch Mean": -0.20977401733398438, | |
| "accuracy": 0.8515625, | |
| "epoch": 0.1825, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.185, | |
| "grad_norm": 7.816598892211914, | |
| "learning_rate": 2.573684210526316e-06, | |
| "loss": 0.4101, | |
| "step": 74 | |
| }, | |
| { | |
| "Batch Mean": -0.6599991321563721, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.185, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 13.089007377624512, | |
| "learning_rate": 2.5657894736842107e-06, | |
| "loss": 0.427, | |
| "step": 75 | |
| }, | |
| { | |
| "Batch Mean": -0.37617337703704834, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.1875, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 12.075210571289062, | |
| "learning_rate": 2.5578947368421054e-06, | |
| "loss": 0.4797, | |
| "step": 76 | |
| }, | |
| { | |
| "Batch Mean": 0.1281442642211914, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.19, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.1925, | |
| "grad_norm": 9.70462703704834, | |
| "learning_rate": 2.55e-06, | |
| "loss": 0.4425, | |
| "step": 77 | |
| }, | |
| { | |
| "Batch Mean": 0.6071650981903076, | |
| "accuracy": 0.828125, | |
| "epoch": 0.1925, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.195, | |
| "grad_norm": 14.395353317260742, | |
| "learning_rate": 2.542105263157895e-06, | |
| "loss": 0.4016, | |
| "step": 78 | |
| }, | |
| { | |
| "Batch Mean": 0.34730714559555054, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.195, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1975, | |
| "grad_norm": 9.122536659240723, | |
| "learning_rate": 2.5342105263157892e-06, | |
| "loss": 0.3994, | |
| "step": 79 | |
| }, | |
| { | |
| "Batch Mean": 0.36483922600746155, | |
| "accuracy": 0.828125, | |
| "epoch": 0.1975, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 10.289307594299316, | |
| "learning_rate": 2.526315789473684e-06, | |
| "loss": 0.368, | |
| "step": 80 | |
| }, | |
| { | |
| "Batch Mean": -0.20039799809455872, | |
| "accuracy": 0.8125, | |
| "epoch": 0.2, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2025, | |
| "grad_norm": 7.813342094421387, | |
| "learning_rate": 2.5184210526315787e-06, | |
| "loss": 0.3679, | |
| "step": 81 | |
| }, | |
| { | |
| "Batch Mean": -0.1629079282283783, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.2025, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.205, | |
| "grad_norm": 7.504952430725098, | |
| "learning_rate": 2.510526315789474e-06, | |
| "loss": 0.3823, | |
| "step": 82 | |
| }, | |
| { | |
| "Batch Mean": -0.07863587141036987, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.205, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.2075, | |
| "grad_norm": 7.721461296081543, | |
| "learning_rate": 2.5026315789473686e-06, | |
| "loss": 0.3967, | |
| "step": 83 | |
| }, | |
| { | |
| "Batch Mean": 0.17194491624832153, | |
| "accuracy": 0.8125, | |
| "epoch": 0.2075, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 8.062063217163086, | |
| "learning_rate": 2.4947368421052634e-06, | |
| "loss": 0.4195, | |
| "step": 84 | |
| }, | |
| { | |
| "Batch Mean": 0.11748838424682617, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.21, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.2125, | |
| "grad_norm": 8.320355415344238, | |
| "learning_rate": 2.486842105263158e-06, | |
| "loss": 0.3864, | |
| "step": 85 | |
| }, | |
| { | |
| "Batch Mean": -0.40801382064819336, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.2125, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.215, | |
| "grad_norm": 14.8066987991333, | |
| "learning_rate": 2.478947368421053e-06, | |
| "loss": 0.4639, | |
| "step": 86 | |
| }, | |
| { | |
| "Batch Mean": -0.35811758041381836, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.215, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.2175, | |
| "grad_norm": 14.074177742004395, | |
| "learning_rate": 2.4710526315789476e-06, | |
| "loss": 0.4999, | |
| "step": 87 | |
| }, | |
| { | |
| "Batch Mean": 0.278584361076355, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.2175, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 10.59004020690918, | |
| "learning_rate": 2.4631578947368424e-06, | |
| "loss": 0.404, | |
| "step": 88 | |
| }, | |
| { | |
| "Batch Mean": -0.21054387092590332, | |
| "accuracy": 0.84375, | |
| "epoch": 0.22, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.2225, | |
| "grad_norm": 8.488351821899414, | |
| "learning_rate": 2.4552631578947367e-06, | |
| "loss": 0.3517, | |
| "step": 89 | |
| }, | |
| { | |
| "Batch Mean": -0.22967231273651123, | |
| "accuracy": 0.84375, | |
| "epoch": 0.2225, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 10.347502708435059, | |
| "learning_rate": 2.4473684210526314e-06, | |
| "loss": 0.3382, | |
| "step": 90 | |
| }, | |
| { | |
| "Batch Mean": 0.1892259418964386, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.225, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2275, | |
| "grad_norm": 10.092534065246582, | |
| "learning_rate": 2.439473684210526e-06, | |
| "loss": 0.4586, | |
| "step": 91 | |
| }, | |
| { | |
| "Batch Mean": 0.516247570514679, | |
| "accuracy": 0.8828125, | |
| "epoch": 0.2275, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 13.944212913513184, | |
| "learning_rate": 2.431578947368421e-06, | |
| "loss": 0.3067, | |
| "step": 92 | |
| }, | |
| { | |
| "Batch Mean": 0.056681275367736816, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.23, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2325, | |
| "grad_norm": 8.53646183013916, | |
| "learning_rate": 2.4236842105263157e-06, | |
| "loss": 0.3946, | |
| "step": 93 | |
| }, | |
| { | |
| "Batch Mean": -0.47129684686660767, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.2325, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.235, | |
| "grad_norm": 11.441537857055664, | |
| "learning_rate": 2.4157894736842104e-06, | |
| "loss": 0.3891, | |
| "step": 94 | |
| }, | |
| { | |
| "Batch Mean": -0.3262195587158203, | |
| "accuracy": 0.84375, | |
| "epoch": 0.235, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.2375, | |
| "grad_norm": 10.751629829406738, | |
| "learning_rate": 2.4078947368421056e-06, | |
| "loss": 0.3578, | |
| "step": 95 | |
| }, | |
| { | |
| "Batch Mean": -0.6593484878540039, | |
| "accuracy": 0.828125, | |
| "epoch": 0.2375, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 15.914350509643555, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.4204, | |
| "step": 96 | |
| }, | |
| { | |
| "Batch Mean": -0.32147085666656494, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.24, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2425, | |
| "grad_norm": 11.814691543579102, | |
| "learning_rate": 2.392105263157895e-06, | |
| "loss": 0.3892, | |
| "step": 97 | |
| }, | |
| { | |
| "Batch Mean": 0.49291136860847473, | |
| "accuracy": 0.875, | |
| "epoch": 0.2425, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.245, | |
| "grad_norm": 11.86365795135498, | |
| "learning_rate": 2.38421052631579e-06, | |
| "loss": 0.3264, | |
| "step": 98 | |
| }, | |
| { | |
| "Batch Mean": 0.49174070358276367, | |
| "accuracy": 0.8515625, | |
| "epoch": 0.245, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.2475, | |
| "grad_norm": 10.874544143676758, | |
| "learning_rate": 2.376315789473684e-06, | |
| "loss": 0.3847, | |
| "step": 99 | |
| }, | |
| { | |
| "Batch Mean": 0.6919900178909302, | |
| "accuracy": 0.796875, | |
| "epoch": 0.2475, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 15.105185508728027, | |
| "learning_rate": 2.368421052631579e-06, | |
| "loss": 0.4066, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |