| { | |
| "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": -3.4444580078125, | |
| "accuracy": 0.40625, | |
| "epoch": 0, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0.0025, | |
| "grad_norm": 43.31455612182617, | |
| "learning_rate": 1.5000000000000002e-07, | |
| "loss": 0.9077, | |
| "step": 1 | |
| }, | |
| { | |
| "Batch Mean": -3.43255615234375, | |
| "accuracy": 0.4921875, | |
| "epoch": 0.0025, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005, | |
| "grad_norm": 50.06682205200195, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 0.8907, | |
| "step": 2 | |
| }, | |
| { | |
| "Batch Mean": -3.44775390625, | |
| "accuracy": 0.453125, | |
| "epoch": 0.005, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0075, | |
| "grad_norm": 47.389522552490234, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.8899, | |
| "step": 3 | |
| }, | |
| { | |
| "Batch Mean": -3.4224853515625, | |
| "accuracy": 0.4375, | |
| "epoch": 0.0075, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 49.24280548095703, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 0.8833, | |
| "step": 4 | |
| }, | |
| { | |
| "Batch Mean": -3.39697265625, | |
| "accuracy": 0.5234375, | |
| "epoch": 0.01, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0125, | |
| "grad_norm": 44.11930847167969, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.8661, | |
| "step": 5 | |
| }, | |
| { | |
| "Batch Mean": -3.38104248046875, | |
| "accuracy": 0.5078125, | |
| "epoch": 0.0125, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015, | |
| "grad_norm": 44.9741325378418, | |
| "learning_rate": 9e-07, | |
| "loss": 0.8805, | |
| "step": 6 | |
| }, | |
| { | |
| "Batch Mean": -3.19671630859375, | |
| "accuracy": 0.5078125, | |
| "epoch": 0.015, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0175, | |
| "grad_norm": 42.32276153564453, | |
| "learning_rate": 1.05e-06, | |
| "loss": 0.8547, | |
| "step": 7 | |
| }, | |
| { | |
| "Batch Mean": -3.1279296875, | |
| "accuracy": 0.4921875, | |
| "epoch": 0.0175, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 44.22520065307617, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 0.8464, | |
| "step": 8 | |
| }, | |
| { | |
| "Batch Mean": -2.520538330078125, | |
| "accuracy": 0.5, | |
| "epoch": 0.02, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0225, | |
| "grad_norm": 36.41325378417969, | |
| "learning_rate": 1.35e-06, | |
| "loss": 0.7946, | |
| "step": 9 | |
| }, | |
| { | |
| "Batch Mean": -2.413787841796875, | |
| "accuracy": 0.515625, | |
| "epoch": 0.0225, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 31.74874496459961, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.7939, | |
| "step": 10 | |
| }, | |
| { | |
| "Batch Mean": -2.218231201171875, | |
| "accuracy": 0.4921875, | |
| "epoch": 0.025, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0275, | |
| "grad_norm": 34.12691879272461, | |
| "learning_rate": 1.65e-06, | |
| "loss": 0.7505, | |
| "step": 11 | |
| }, | |
| { | |
| "Batch Mean": -0.2393415868282318, | |
| "accuracy": 0.53125, | |
| "epoch": 0.0275, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 10.481990814208984, | |
| "learning_rate": 1.8e-06, | |
| "loss": 0.6779, | |
| "step": 12 | |
| }, | |
| { | |
| "Batch Mean": 0.2733480930328369, | |
| "accuracy": 0.5703125, | |
| "epoch": 0.03, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0325, | |
| "grad_norm": 8.39962387084961, | |
| "learning_rate": 1.95e-06, | |
| "loss": 0.6728, | |
| "step": 13 | |
| }, | |
| { | |
| "Batch Mean": 0.5041149854660034, | |
| "accuracy": 0.671875, | |
| "epoch": 0.0325, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.035, | |
| "grad_norm": 15.906760215759277, | |
| "learning_rate": 2.1e-06, | |
| "loss": 0.6679, | |
| "step": 14 | |
| }, | |
| { | |
| "Batch Mean": 0.8821840286254883, | |
| "accuracy": 0.578125, | |
| "epoch": 0.035, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0375, | |
| "grad_norm": 19.03310775756836, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.7024, | |
| "step": 15 | |
| }, | |
| { | |
| "Batch Mean": 1.1891746520996094, | |
| "accuracy": 0.5859375, | |
| "epoch": 0.0375, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 20.89366340637207, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.6868, | |
| "step": 16 | |
| }, | |
| { | |
| "Batch Mean": 2.1533203125, | |
| "accuracy": 0.671875, | |
| "epoch": 0.04, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0425, | |
| "grad_norm": 34.633541107177734, | |
| "learning_rate": 2.55e-06, | |
| "loss": 0.7297, | |
| "step": 17 | |
| }, | |
| { | |
| "Batch Mean": 2.5652008056640625, | |
| "accuracy": 0.6484375, | |
| "epoch": 0.0425, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.045, | |
| "grad_norm": 41.46820068359375, | |
| "learning_rate": 2.7e-06, | |
| "loss": 0.7387, | |
| "step": 18 | |
| }, | |
| { | |
| "Batch Mean": 2.4411849975585938, | |
| "accuracy": 0.75, | |
| "epoch": 0.045, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0475, | |
| "grad_norm": 42.9193000793457, | |
| "learning_rate": 2.85e-06, | |
| "loss": 0.6813, | |
| "step": 19 | |
| }, | |
| { | |
| "Batch Mean": 2.1285018920898438, | |
| "accuracy": 0.703125, | |
| "epoch": 0.0475, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 37.16952133178711, | |
| "learning_rate": 3e-06, | |
| "loss": 0.672, | |
| "step": 20 | |
| }, | |
| { | |
| "Batch Mean": 1.4230481386184692, | |
| "accuracy": 0.671875, | |
| "epoch": 0.05, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0525, | |
| "grad_norm": 27.2924747467041, | |
| "learning_rate": 2.992105263157895e-06, | |
| "loss": 0.647, | |
| "step": 21 | |
| }, | |
| { | |
| "Batch Mean": 0.5414033532142639, | |
| "accuracy": 0.6953125, | |
| "epoch": 0.0525, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.055, | |
| "grad_norm": 11.927323341369629, | |
| "learning_rate": 2.9842105263157896e-06, | |
| "loss": 0.5981, | |
| "step": 22 | |
| }, | |
| { | |
| "Batch Mean": -0.36795324087142944, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.055, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0575, | |
| "grad_norm": 13.455949783325195, | |
| "learning_rate": 2.9763157894736843e-06, | |
| "loss": 0.5763, | |
| "step": 23 | |
| }, | |
| { | |
| "Batch Mean": -1.0874066352844238, | |
| "accuracy": 0.7421875, | |
| "epoch": 0.0575, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 22.382274627685547, | |
| "learning_rate": 2.968421052631579e-06, | |
| "loss": 0.5763, | |
| "step": 24 | |
| }, | |
| { | |
| "Batch Mean": -1.103028416633606, | |
| "accuracy": 0.6953125, | |
| "epoch": 0.06, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 19.854228973388672, | |
| "learning_rate": 2.960526315789474e-06, | |
| "loss": 0.6065, | |
| "step": 25 | |
| }, | |
| { | |
| "Batch Mean": -0.8232302069664001, | |
| "accuracy": 0.734375, | |
| "epoch": 0.0625, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.065, | |
| "grad_norm": 16.642824172973633, | |
| "learning_rate": 2.9526315789473685e-06, | |
| "loss": 0.5282, | |
| "step": 26 | |
| }, | |
| { | |
| "Batch Mean": -0.11966800689697266, | |
| "accuracy": 0.6875, | |
| "epoch": 0.065, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0675, | |
| "grad_norm": 8.563494682312012, | |
| "learning_rate": 2.9447368421052633e-06, | |
| "loss": 0.5348, | |
| "step": 27 | |
| }, | |
| { | |
| "Batch Mean": 0.5674030780792236, | |
| "accuracy": 0.671875, | |
| "epoch": 0.0675, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 13.211753845214844, | |
| "learning_rate": 2.936842105263158e-06, | |
| "loss": 0.5812, | |
| "step": 28 | |
| }, | |
| { | |
| "Batch Mean": 1.1245126724243164, | |
| "accuracy": 0.6171875, | |
| "epoch": 0.07, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0725, | |
| "grad_norm": 22.8548583984375, | |
| "learning_rate": 2.9289473684210528e-06, | |
| "loss": 0.7336, | |
| "step": 29 | |
| }, | |
| { | |
| "Batch Mean": 1.2212319374084473, | |
| "accuracy": 0.6796875, | |
| "epoch": 0.0725, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 21.96833038330078, | |
| "learning_rate": 2.9210526315789475e-06, | |
| "loss": 0.6167, | |
| "step": 30 | |
| }, | |
| { | |
| "Batch Mean": 0.7509702444076538, | |
| "accuracy": 0.796875, | |
| "epoch": 0.075, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0775, | |
| "grad_norm": 12.47424030303955, | |
| "learning_rate": 2.9131578947368423e-06, | |
| "loss": 0.463, | |
| "step": 31 | |
| }, | |
| { | |
| "Batch Mean": 0.2412339150905609, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.0775, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 9.769195556640625, | |
| "learning_rate": 2.905263157894737e-06, | |
| "loss": 0.5412, | |
| "step": 32 | |
| }, | |
| { | |
| "Batch Mean": -0.5602660179138184, | |
| "accuracy": 0.7109375, | |
| "epoch": 0.08, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0825, | |
| "grad_norm": 11.323891639709473, | |
| "learning_rate": 2.8973684210526318e-06, | |
| "loss": 0.5601, | |
| "step": 33 | |
| }, | |
| { | |
| "Batch Mean": -0.8669416904449463, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.0825, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.085, | |
| "grad_norm": 15.461758613586426, | |
| "learning_rate": 2.8894736842105265e-06, | |
| "loss": 0.5147, | |
| "step": 34 | |
| }, | |
| { | |
| "Batch Mean": -0.7635477781295776, | |
| "accuracy": 0.75, | |
| "epoch": 0.085, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0875, | |
| "grad_norm": 14.736496925354004, | |
| "learning_rate": 2.8815789473684213e-06, | |
| "loss": 0.5421, | |
| "step": 35 | |
| }, | |
| { | |
| "Batch Mean": -0.5088728070259094, | |
| "accuracy": 0.734375, | |
| "epoch": 0.0875, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 11.982501029968262, | |
| "learning_rate": 2.873684210526316e-06, | |
| "loss": 0.5234, | |
| "step": 36 | |
| }, | |
| { | |
| "Batch Mean": -0.03705340623855591, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.09, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0925, | |
| "grad_norm": 7.854848861694336, | |
| "learning_rate": 2.8657894736842103e-06, | |
| "loss": 0.4996, | |
| "step": 37 | |
| }, | |
| { | |
| "Batch Mean": 0.5426892042160034, | |
| "accuracy": 0.734375, | |
| "epoch": 0.0925, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.095, | |
| "grad_norm": 10.834790229797363, | |
| "learning_rate": 2.857894736842105e-06, | |
| "loss": 0.5187, | |
| "step": 38 | |
| }, | |
| { | |
| "Batch Mean": 0.4622654914855957, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.095, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0975, | |
| "grad_norm": 10.764225959777832, | |
| "learning_rate": 2.85e-06, | |
| "loss": 0.4723, | |
| "step": 39 | |
| }, | |
| { | |
| "Batch Mean": 0.18012571334838867, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.0975, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 9.500924110412598, | |
| "learning_rate": 2.8421052631578946e-06, | |
| "loss": 0.4545, | |
| "step": 40 | |
| }, | |
| { | |
| "Batch Mean": -0.05992317199707031, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.1, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1025, | |
| "grad_norm": 8.079994201660156, | |
| "learning_rate": 2.8342105263157897e-06, | |
| "loss": 0.49, | |
| "step": 41 | |
| }, | |
| { | |
| "Batch Mean": 0.04392993450164795, | |
| "accuracy": 0.8515625, | |
| "epoch": 0.1025, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.105, | |
| "grad_norm": 10.184918403625488, | |
| "learning_rate": 2.8263157894736845e-06, | |
| "loss": 0.3789, | |
| "step": 42 | |
| }, | |
| { | |
| "Batch Mean": -0.2515716552734375, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.105, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1075, | |
| "grad_norm": 9.894248008728027, | |
| "learning_rate": 2.8184210526315792e-06, | |
| "loss": 0.435, | |
| "step": 43 | |
| }, | |
| { | |
| "Batch Mean": 0.15310335159301758, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.1075, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 10.671575546264648, | |
| "learning_rate": 2.810526315789474e-06, | |
| "loss": 0.4291, | |
| "step": 44 | |
| }, | |
| { | |
| "Batch Mean": 0.2563772201538086, | |
| "accuracy": 0.765625, | |
| "epoch": 0.11, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1125, | |
| "grad_norm": 10.690945625305176, | |
| "learning_rate": 2.8026315789473687e-06, | |
| "loss": 0.4812, | |
| "step": 45 | |
| }, | |
| { | |
| "Batch Mean": 0.046463966369628906, | |
| "accuracy": 0.875, | |
| "epoch": 0.1125, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.115, | |
| "grad_norm": 9.616545677185059, | |
| "learning_rate": 2.7947368421052635e-06, | |
| "loss": 0.3082, | |
| "step": 46 | |
| }, | |
| { | |
| "Batch Mean": 0.11115455627441406, | |
| "accuracy": 0.75, | |
| "epoch": 0.115, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.1175, | |
| "grad_norm": 15.726456642150879, | |
| "learning_rate": 2.7868421052631578e-06, | |
| "loss": 0.5576, | |
| "step": 47 | |
| }, | |
| { | |
| "Batch Mean": 0.2878689765930176, | |
| "accuracy": 0.8125, | |
| "epoch": 0.1175, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 11.675898551940918, | |
| "learning_rate": 2.7789473684210525e-06, | |
| "loss": 0.472, | |
| "step": 48 | |
| }, | |
| { | |
| "Batch Mean": -0.25928783416748047, | |
| "accuracy": 0.75, | |
| "epoch": 0.12, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1225, | |
| "grad_norm": 15.394309997558594, | |
| "learning_rate": 2.7710526315789473e-06, | |
| "loss": 0.5011, | |
| "step": 49 | |
| }, | |
| { | |
| "Batch Mean": -0.2885439097881317, | |
| "accuracy": 0.703125, | |
| "epoch": 0.1225, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 15.69314956665039, | |
| "learning_rate": 2.763157894736842e-06, | |
| "loss": 0.5974, | |
| "step": 50 | |
| }, | |
| { | |
| "Batch Mean": -0.8389434814453125, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.125, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1275, | |
| "grad_norm": 16.744606018066406, | |
| "learning_rate": 2.7552631578947368e-06, | |
| "loss": 0.5151, | |
| "step": 51 | |
| }, | |
| { | |
| "Batch Mean": -0.6685854196548462, | |
| "accuracy": 0.703125, | |
| "epoch": 0.1275, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 15.530059814453125, | |
| "learning_rate": 2.7473684210526315e-06, | |
| "loss": 0.5502, | |
| "step": 52 | |
| }, | |
| { | |
| "Batch Mean": -0.5919044017791748, | |
| "accuracy": 0.71875, | |
| "epoch": 0.13, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.1325, | |
| "grad_norm": 15.14562702178955, | |
| "learning_rate": 2.7394736842105263e-06, | |
| "loss": 0.5305, | |
| "step": 53 | |
| }, | |
| { | |
| "Batch Mean": -0.11876988410949707, | |
| "accuracy": 0.75, | |
| "epoch": 0.1325, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.135, | |
| "grad_norm": 7.011972427368164, | |
| "learning_rate": 2.7315789473684214e-06, | |
| "loss": 0.4802, | |
| "step": 54 | |
| }, | |
| { | |
| "Batch Mean": 0.09675300121307373, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.135, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1375, | |
| "grad_norm": 8.37578010559082, | |
| "learning_rate": 2.723684210526316e-06, | |
| "loss": 0.4354, | |
| "step": 55 | |
| }, | |
| { | |
| "Batch Mean": 0.3877286911010742, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1375, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 8.380095481872559, | |
| "learning_rate": 2.715789473684211e-06, | |
| "loss": 0.4114, | |
| "step": 56 | |
| }, | |
| { | |
| "Batch Mean": 0.7472854852676392, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.14, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1425, | |
| "grad_norm": 12.794017791748047, | |
| "learning_rate": 2.7078947368421052e-06, | |
| "loss": 0.465, | |
| "step": 57 | |
| }, | |
| { | |
| "Batch Mean": 0.5131087303161621, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1425, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.145, | |
| "grad_norm": 8.367005348205566, | |
| "learning_rate": 2.7e-06, | |
| "loss": 0.4407, | |
| "step": 58 | |
| }, | |
| { | |
| "Batch Mean": 0.5741372108459473, | |
| "accuracy": 0.8515625, | |
| "epoch": 0.145, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.1475, | |
| "grad_norm": 9.066951751708984, | |
| "learning_rate": 2.6921052631578947e-06, | |
| "loss": 0.4468, | |
| "step": 59 | |
| }, | |
| { | |
| "Batch Mean": 0.169053316116333, | |
| "accuracy": 0.734375, | |
| "epoch": 0.1475, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 6.888144016265869, | |
| "learning_rate": 2.6842105263157895e-06, | |
| "loss": 0.4849, | |
| "step": 60 | |
| }, | |
| { | |
| "Batch Mean": -0.18010663986206055, | |
| "accuracy": 0.765625, | |
| "epoch": 0.15, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1525, | |
| "grad_norm": 6.846613883972168, | |
| "learning_rate": 2.6763157894736842e-06, | |
| "loss": 0.4643, | |
| "step": 61 | |
| }, | |
| { | |
| "Batch Mean": -0.4085197448730469, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.1525, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.155, | |
| "grad_norm": 7.677695274353027, | |
| "learning_rate": 2.668421052631579e-06, | |
| "loss": 0.3926, | |
| "step": 62 | |
| }, | |
| { | |
| "Batch Mean": -0.48792409896850586, | |
| "accuracy": 0.828125, | |
| "epoch": 0.155, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.1575, | |
| "grad_norm": 8.656214714050293, | |
| "learning_rate": 2.6605263157894737e-06, | |
| "loss": 0.4116, | |
| "step": 63 | |
| }, | |
| { | |
| "Batch Mean": -0.19100165367126465, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.1575, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 9.927976608276367, | |
| "learning_rate": 2.6526315789473685e-06, | |
| "loss": 0.4395, | |
| "step": 64 | |
| }, | |
| { | |
| "Batch Mean": 0.025106430053710938, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.16, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1625, | |
| "grad_norm": 8.97470474243164, | |
| "learning_rate": 2.644736842105263e-06, | |
| "loss": 0.4632, | |
| "step": 65 | |
| }, | |
| { | |
| "Batch Mean": 0.4497801661491394, | |
| "accuracy": 0.796875, | |
| "epoch": 0.1625, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.165, | |
| "grad_norm": 11.160425186157227, | |
| "learning_rate": 2.636842105263158e-06, | |
| "loss": 0.5073, | |
| "step": 66 | |
| }, | |
| { | |
| "Batch Mean": 0.6494235992431641, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.165, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.1675, | |
| "grad_norm": 11.464187622070312, | |
| "learning_rate": 2.6289473684210527e-06, | |
| "loss": 0.4467, | |
| "step": 67 | |
| }, | |
| { | |
| "Batch Mean": -0.17097759246826172, | |
| "accuracy": 0.828125, | |
| "epoch": 0.1675, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 8.820326805114746, | |
| "learning_rate": 2.6210526315789474e-06, | |
| "loss": 0.4543, | |
| "step": 68 | |
| }, | |
| { | |
| "Batch Mean": -0.1554449200630188, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.17, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.1725, | |
| "grad_norm": 9.248348236083984, | |
| "learning_rate": 2.613157894736842e-06, | |
| "loss": 0.3796, | |
| "step": 69 | |
| }, | |
| { | |
| "Batch Mean": -0.21274375915527344, | |
| "accuracy": 0.75, | |
| "epoch": 0.1725, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 8.15241813659668, | |
| "learning_rate": 2.605263157894737e-06, | |
| "loss": 0.4445, | |
| "step": 70 | |
| }, | |
| { | |
| "Batch Mean": -0.06371593475341797, | |
| "accuracy": 0.765625, | |
| "epoch": 0.175, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1775, | |
| "grad_norm": 6.543805122375488, | |
| "learning_rate": 2.5973684210526317e-06, | |
| "loss": 0.4376, | |
| "step": 71 | |
| }, | |
| { | |
| "Batch Mean": -0.10743597149848938, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.1775, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 7.594844341278076, | |
| "learning_rate": 2.5894736842105264e-06, | |
| "loss": 0.4363, | |
| "step": 72 | |
| }, | |
| { | |
| "Batch Mean": 0.08045387268066406, | |
| "accuracy": 0.84375, | |
| "epoch": 0.18, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1825, | |
| "grad_norm": 7.977687835693359, | |
| "learning_rate": 2.581578947368421e-06, | |
| "loss": 0.3616, | |
| "step": 73 | |
| }, | |
| { | |
| "Batch Mean": -0.11765336990356445, | |
| "accuracy": 0.796875, | |
| "epoch": 0.1825, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.185, | |
| "grad_norm": 6.521571636199951, | |
| "learning_rate": 2.573684210526316e-06, | |
| "loss": 0.4001, | |
| "step": 74 | |
| }, | |
| { | |
| "Batch Mean": 0.0607762336730957, | |
| "accuracy": 0.703125, | |
| "epoch": 0.185, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 8.15623950958252, | |
| "learning_rate": 2.5657894736842107e-06, | |
| "loss": 0.4941, | |
| "step": 75 | |
| }, | |
| { | |
| "Batch Mean": 0.4910193681716919, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1875, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 9.602405548095703, | |
| "learning_rate": 2.5578947368421054e-06, | |
| "loss": 0.4224, | |
| "step": 76 | |
| }, | |
| { | |
| "Batch Mean": 0.06563162803649902, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.19, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.1925, | |
| "grad_norm": 6.577213764190674, | |
| "learning_rate": 2.55e-06, | |
| "loss": 0.4047, | |
| "step": 77 | |
| }, | |
| { | |
| "Batch Mean": -0.08942317962646484, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1925, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.195, | |
| "grad_norm": 16.202590942382812, | |
| "learning_rate": 2.542105263157895e-06, | |
| "loss": 0.4418, | |
| "step": 78 | |
| }, | |
| { | |
| "Batch Mean": -0.37958288192749023, | |
| "accuracy": 0.8125, | |
| "epoch": 0.195, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1975, | |
| "grad_norm": 23.45298194885254, | |
| "learning_rate": 2.5342105263157892e-06, | |
| "loss": 0.391, | |
| "step": 79 | |
| }, | |
| { | |
| "Batch Mean": -0.5664916038513184, | |
| "accuracy": 0.8125, | |
| "epoch": 0.1975, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 10.679176330566406, | |
| "learning_rate": 2.526315789473684e-06, | |
| "loss": 0.4402, | |
| "step": 80 | |
| }, | |
| { | |
| "Batch Mean": -0.31339550018310547, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.2, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2025, | |
| "grad_norm": 16.459426879882812, | |
| "learning_rate": 2.5184210526315787e-06, | |
| "loss": 0.3765, | |
| "step": 81 | |
| }, | |
| { | |
| "Batch Mean": 0.20376300811767578, | |
| "accuracy": 0.78125, | |
| "epoch": 0.2025, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.205, | |
| "grad_norm": 8.887246131896973, | |
| "learning_rate": 2.510526315789474e-06, | |
| "loss": 0.4509, | |
| "step": 82 | |
| }, | |
| { | |
| "Batch Mean": -0.054105013608932495, | |
| "accuracy": 0.796875, | |
| "epoch": 0.205, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.2075, | |
| "grad_norm": 8.116570472717285, | |
| "learning_rate": 2.5026315789473686e-06, | |
| "loss": 0.3906, | |
| "step": 83 | |
| }, | |
| { | |
| "Batch Mean": 0.4383058547973633, | |
| "accuracy": 0.875, | |
| "epoch": 0.2075, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 9.300606727600098, | |
| "learning_rate": 2.4947368421052634e-06, | |
| "loss": 0.3099, | |
| "step": 84 | |
| }, | |
| { | |
| "Batch Mean": 0.4242062568664551, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.21, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.2125, | |
| "grad_norm": 8.7870512008667, | |
| "learning_rate": 2.486842105263158e-06, | |
| "loss": 0.3996, | |
| "step": 85 | |
| }, | |
| { | |
| "Batch Mean": 0.22194945812225342, | |
| "accuracy": 0.78125, | |
| "epoch": 0.2125, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.215, | |
| "grad_norm": 8.691889762878418, | |
| "learning_rate": 2.478947368421053e-06, | |
| "loss": 0.4512, | |
| "step": 86 | |
| }, | |
| { | |
| "Batch Mean": -0.14107918739318848, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.215, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.2175, | |
| "grad_norm": 7.0445146560668945, | |
| "learning_rate": 2.4710526315789476e-06, | |
| "loss": 0.3771, | |
| "step": 87 | |
| }, | |
| { | |
| "Batch Mean": -0.35906583070755005, | |
| "accuracy": 0.765625, | |
| "epoch": 0.2175, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 9.609421730041504, | |
| "learning_rate": 2.4631578947368424e-06, | |
| "loss": 0.4558, | |
| "step": 88 | |
| }, | |
| { | |
| "Batch Mean": -0.45313501358032227, | |
| "accuracy": 0.828125, | |
| "epoch": 0.22, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.2225, | |
| "grad_norm": 9.091865539550781, | |
| "learning_rate": 2.4552631578947367e-06, | |
| "loss": 0.4292, | |
| "step": 89 | |
| }, | |
| { | |
| "Batch Mean": -0.35767626762390137, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.2225, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 7.811399459838867, | |
| "learning_rate": 2.4473684210526314e-06, | |
| "loss": 0.3823, | |
| "step": 90 | |
| }, | |
| { | |
| "Batch Mean": 0.0022423267364501953, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.225, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2275, | |
| "grad_norm": 7.804200172424316, | |
| "learning_rate": 2.439473684210526e-06, | |
| "loss": 0.4079, | |
| "step": 91 | |
| }, | |
| { | |
| "Batch Mean": 0.17678523063659668, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.2275, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 7.715360641479492, | |
| "learning_rate": 2.431578947368421e-06, | |
| "loss": 0.391, | |
| "step": 92 | |
| }, | |
| { | |
| "Batch Mean": -0.22452354431152344, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.23, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2325, | |
| "grad_norm": 7.197687149047852, | |
| "learning_rate": 2.4236842105263157e-06, | |
| "loss": 0.3864, | |
| "step": 93 | |
| }, | |
| { | |
| "Batch Mean": 0.30264759063720703, | |
| "accuracy": 0.84375, | |
| "epoch": 0.2325, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.235, | |
| "grad_norm": 7.754411697387695, | |
| "learning_rate": 2.4157894736842104e-06, | |
| "loss": 0.3939, | |
| "step": 94 | |
| }, | |
| { | |
| "Batch Mean": -0.2516747713088989, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.235, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.2375, | |
| "grad_norm": 8.55639362335205, | |
| "learning_rate": 2.4078947368421056e-06, | |
| "loss": 0.4671, | |
| "step": 95 | |
| }, | |
| { | |
| "Batch Mean": -0.15928316116333008, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.2375, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 7.463867664337158, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 0.471, | |
| "step": 96 | |
| }, | |
| { | |
| "Batch Mean": -0.0638132095336914, | |
| "accuracy": 0.8125, | |
| "epoch": 0.24, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2425, | |
| "grad_norm": 6.797260761260986, | |
| "learning_rate": 2.392105263157895e-06, | |
| "loss": 0.3892, | |
| "step": 97 | |
| }, | |
| { | |
| "Batch Mean": -0.08082282543182373, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.2425, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.245, | |
| "grad_norm": 7.096107482910156, | |
| "learning_rate": 2.38421052631579e-06, | |
| "loss": 0.3942, | |
| "step": 98 | |
| }, | |
| { | |
| "Batch Mean": 0.3185189962387085, | |
| "accuracy": 0.796875, | |
| "epoch": 0.245, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.2475, | |
| "grad_norm": 9.447708129882812, | |
| "learning_rate": 2.376315789473684e-06, | |
| "loss": 0.5037, | |
| "step": 99 | |
| }, | |
| { | |
| "Batch Mean": 0.26651012897491455, | |
| "accuracy": 0.890625, | |
| "epoch": 0.2475, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 9.391190528869629, | |
| "learning_rate": 2.368421052631579e-06, | |
| "loss": 0.3352, | |
| "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 | |
| } | |