| { | |
| "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": 0.8096684217453003, | |
| "accuracy": 0.625, | |
| "epoch": 0, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0.0025, | |
| "grad_norm": 9.15977954864502, | |
| "learning_rate": 2.5000000000000004e-07, | |
| "loss": 0.6694, | |
| "step": 1 | |
| }, | |
| { | |
| "Batch Mean": 0.8517913818359375, | |
| "accuracy": 0.5234375, | |
| "epoch": 0.0025, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005, | |
| "grad_norm": 9.550761222839355, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.7001, | |
| "step": 2 | |
| }, | |
| { | |
| "Batch Mean": 0.8251190185546875, | |
| "accuracy": 0.4921875, | |
| "epoch": 0.005, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0075, | |
| "grad_norm": 10.096807479858398, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.7277, | |
| "step": 3 | |
| }, | |
| { | |
| "Batch Mean": 0.8156070709228516, | |
| "accuracy": 0.515625, | |
| "epoch": 0.0075, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 10.845061302185059, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.7116, | |
| "step": 4 | |
| }, | |
| { | |
| "Batch Mean": 0.772618293762207, | |
| "accuracy": 0.53125, | |
| "epoch": 0.01, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0125, | |
| "grad_norm": 10.683115005493164, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.7074, | |
| "step": 5 | |
| }, | |
| { | |
| "Batch Mean": 0.7320594787597656, | |
| "accuracy": 0.5390625, | |
| "epoch": 0.0125, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015, | |
| "grad_norm": 9.102219581604004, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.6964, | |
| "step": 6 | |
| }, | |
| { | |
| "Batch Mean": 0.639228105545044, | |
| "accuracy": 0.5703125, | |
| "epoch": 0.015, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0175, | |
| "grad_norm": 10.126181602478027, | |
| "learning_rate": 1.75e-06, | |
| "loss": 0.6981, | |
| "step": 7 | |
| }, | |
| { | |
| "Batch Mean": 0.28215163946151733, | |
| "accuracy": 0.6015625, | |
| "epoch": 0.0175, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 6.3671040534973145, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.6668, | |
| "step": 8 | |
| }, | |
| { | |
| "Batch Mean": 0.10402120649814606, | |
| "accuracy": 0.5234375, | |
| "epoch": 0.02, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0225, | |
| "grad_norm": 6.053694248199463, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.6907, | |
| "step": 9 | |
| }, | |
| { | |
| "Batch Mean": -0.794627845287323, | |
| "accuracy": 0.5234375, | |
| "epoch": 0.0225, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 9.284210205078125, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.6879, | |
| "step": 10 | |
| }, | |
| { | |
| "Batch Mean": -1.1373445987701416, | |
| "accuracy": 0.59375, | |
| "epoch": 0.025, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0275, | |
| "grad_norm": 13.110421180725098, | |
| "learning_rate": 2.7500000000000004e-06, | |
| "loss": 0.6954, | |
| "step": 11 | |
| }, | |
| { | |
| "Batch Mean": -1.2041501998901367, | |
| "accuracy": 0.59375, | |
| "epoch": 0.0275, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 13.619694709777832, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7004, | |
| "step": 12 | |
| }, | |
| { | |
| "Batch Mean": -1.1390533447265625, | |
| "accuracy": 0.65625, | |
| "epoch": 0.03, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0325, | |
| "grad_norm": 16.698774337768555, | |
| "learning_rate": 3.2500000000000002e-06, | |
| "loss": 0.6691, | |
| "step": 13 | |
| }, | |
| { | |
| "Batch Mean": -1.2006721496582031, | |
| "accuracy": 0.6953125, | |
| "epoch": 0.0325, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.035, | |
| "grad_norm": 13.031152725219727, | |
| "learning_rate": 3.5e-06, | |
| "loss": 0.6749, | |
| "step": 14 | |
| }, | |
| { | |
| "Batch Mean": -1.1018962860107422, | |
| "accuracy": 0.6484375, | |
| "epoch": 0.035, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0375, | |
| "grad_norm": 11.913775444030762, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.6517, | |
| "step": 15 | |
| }, | |
| { | |
| "Batch Mean": -0.8371965289115906, | |
| "accuracy": 0.65625, | |
| "epoch": 0.0375, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 9.776491165161133, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.6435, | |
| "step": 16 | |
| }, | |
| { | |
| "Batch Mean": -0.25420451164245605, | |
| "accuracy": 0.65625, | |
| "epoch": 0.04, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0425, | |
| "grad_norm": 5.094934940338135, | |
| "learning_rate": 4.25e-06, | |
| "loss": 0.6352, | |
| "step": 17 | |
| }, | |
| { | |
| "Batch Mean": 0.45112472772598267, | |
| "accuracy": 0.625, | |
| "epoch": 0.0425, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.045, | |
| "grad_norm": 6.203393459320068, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.6305, | |
| "step": 18 | |
| }, | |
| { | |
| "Batch Mean": 1.0782840251922607, | |
| "accuracy": 0.640625, | |
| "epoch": 0.045, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0475, | |
| "grad_norm": 11.932191848754883, | |
| "learning_rate": 4.75e-06, | |
| "loss": 0.6401, | |
| "step": 19 | |
| }, | |
| { | |
| "Batch Mean": 1.6000666618347168, | |
| "accuracy": 0.65625, | |
| "epoch": 0.0475, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 18.466459274291992, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6928, | |
| "step": 20 | |
| }, | |
| { | |
| "Batch Mean": 1.6292922496795654, | |
| "accuracy": 0.65625, | |
| "epoch": 0.05, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0525, | |
| "grad_norm": 18.958446502685547, | |
| "learning_rate": 4.986842105263158e-06, | |
| "loss": 0.6456, | |
| "step": 21 | |
| }, | |
| { | |
| "Batch Mean": 1.0992889404296875, | |
| "accuracy": 0.7109375, | |
| "epoch": 0.0525, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.055, | |
| "grad_norm": 12.98255443572998, | |
| "learning_rate": 4.973684210526316e-06, | |
| "loss": 0.606, | |
| "step": 22 | |
| }, | |
| { | |
| "Batch Mean": 0.33066248893737793, | |
| "accuracy": 0.703125, | |
| "epoch": 0.055, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0575, | |
| "grad_norm": 6.1022725105285645, | |
| "learning_rate": 4.960526315789474e-06, | |
| "loss": 0.5807, | |
| "step": 23 | |
| }, | |
| { | |
| "Batch Mean": -0.7870486974716187, | |
| "accuracy": 0.7109375, | |
| "epoch": 0.0575, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 9.311738014221191, | |
| "learning_rate": 4.947368421052632e-06, | |
| "loss": 0.5331, | |
| "step": 24 | |
| }, | |
| { | |
| "Batch Mean": -1.2838190793991089, | |
| "accuracy": 0.6484375, | |
| "epoch": 0.06, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 13.955443382263184, | |
| "learning_rate": 4.9342105263157895e-06, | |
| "loss": 0.6186, | |
| "step": 25 | |
| }, | |
| { | |
| "Batch Mean": -1.324391484260559, | |
| "accuracy": 0.6953125, | |
| "epoch": 0.0625, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.065, | |
| "grad_norm": 14.612135887145996, | |
| "learning_rate": 4.921052631578948e-06, | |
| "loss": 0.5325, | |
| "step": 26 | |
| }, | |
| { | |
| "Batch Mean": -0.7885305881500244, | |
| "accuracy": 0.7109375, | |
| "epoch": 0.065, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0675, | |
| "grad_norm": 9.373554229736328, | |
| "learning_rate": 4.907894736842106e-06, | |
| "loss": 0.5549, | |
| "step": 27 | |
| }, | |
| { | |
| "Batch Mean": -0.25717926025390625, | |
| "accuracy": 0.6953125, | |
| "epoch": 0.0675, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 5.559288501739502, | |
| "learning_rate": 4.894736842105264e-06, | |
| "loss": 0.572, | |
| "step": 28 | |
| }, | |
| { | |
| "Batch Mean": 0.7993221282958984, | |
| "accuracy": 0.6171875, | |
| "epoch": 0.07, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0725, | |
| "grad_norm": 10.778632164001465, | |
| "learning_rate": 4.881578947368422e-06, | |
| "loss": 0.6806, | |
| "step": 29 | |
| }, | |
| { | |
| "Batch Mean": 1.2803305387496948, | |
| "accuracy": 0.6796875, | |
| "epoch": 0.0725, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 14.314054489135742, | |
| "learning_rate": 4.8684210526315795e-06, | |
| "loss": 0.641, | |
| "step": 30 | |
| }, | |
| { | |
| "Batch Mean": 1.1847546100616455, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.075, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0775, | |
| "grad_norm": 12.86508846282959, | |
| "learning_rate": 4.855263157894737e-06, | |
| "loss": 0.5118, | |
| "step": 31 | |
| }, | |
| { | |
| "Batch Mean": 0.7971279621124268, | |
| "accuracy": 0.734375, | |
| "epoch": 0.0775, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 8.987775802612305, | |
| "learning_rate": 4.842105263157895e-06, | |
| "loss": 0.5508, | |
| "step": 32 | |
| }, | |
| { | |
| "Batch Mean": -0.12238264083862305, | |
| "accuracy": 0.671875, | |
| "epoch": 0.08, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0825, | |
| "grad_norm": 4.265674591064453, | |
| "learning_rate": 4.828947368421053e-06, | |
| "loss": 0.5533, | |
| "step": 33 | |
| }, | |
| { | |
| "Batch Mean": -0.7733749151229858, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.0825, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.085, | |
| "grad_norm": 9.074788093566895, | |
| "learning_rate": 4.815789473684211e-06, | |
| "loss": 0.5455, | |
| "step": 34 | |
| }, | |
| { | |
| "Batch Mean": -0.9033082723617554, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.085, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0875, | |
| "grad_norm": 10.331607818603516, | |
| "learning_rate": 4.802631578947369e-06, | |
| "loss": 0.5539, | |
| "step": 35 | |
| }, | |
| { | |
| "Batch Mean": -0.5780456066131592, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.0875, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 7.254377365112305, | |
| "learning_rate": 4.789473684210527e-06, | |
| "loss": 0.547, | |
| "step": 36 | |
| }, | |
| { | |
| "Batch Mean": 0.09664157032966614, | |
| "accuracy": 0.6953125, | |
| "epoch": 0.09, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0925, | |
| "grad_norm": 3.9853391647338867, | |
| "learning_rate": 4.7763157894736844e-06, | |
| "loss": 0.5451, | |
| "step": 37 | |
| }, | |
| { | |
| "Batch Mean": 0.5406360626220703, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.0925, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.095, | |
| "grad_norm": 7.308631896972656, | |
| "learning_rate": 4.763157894736842e-06, | |
| "loss": 0.5313, | |
| "step": 38 | |
| }, | |
| { | |
| "Batch Mean": 0.6023058891296387, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.095, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0975, | |
| "grad_norm": 7.4278717041015625, | |
| "learning_rate": 4.75e-06, | |
| "loss": 0.5145, | |
| "step": 39 | |
| }, | |
| { | |
| "Batch Mean": 0.23680943250656128, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.0975, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 5.057112693786621, | |
| "learning_rate": 4.736842105263158e-06, | |
| "loss": 0.4793, | |
| "step": 40 | |
| }, | |
| { | |
| "Batch Mean": -0.2819175720214844, | |
| "accuracy": 0.75, | |
| "epoch": 0.1, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1025, | |
| "grad_norm": 5.986076831817627, | |
| "learning_rate": 4.723684210526316e-06, | |
| "loss": 0.4989, | |
| "step": 41 | |
| }, | |
| { | |
| "Batch Mean": -0.2952081263065338, | |
| "accuracy": 0.765625, | |
| "epoch": 0.1025, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.105, | |
| "grad_norm": 6.123879432678223, | |
| "learning_rate": 4.710526315789474e-06, | |
| "loss": 0.4959, | |
| "step": 42 | |
| }, | |
| { | |
| "Batch Mean": -0.5327777862548828, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.105, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1075, | |
| "grad_norm": 7.773658752441406, | |
| "learning_rate": 4.697368421052632e-06, | |
| "loss": 0.4501, | |
| "step": 43 | |
| }, | |
| { | |
| "Batch Mean": -0.00808095932006836, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.1075, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 6.165060520172119, | |
| "learning_rate": 4.68421052631579e-06, | |
| "loss": 0.4952, | |
| "step": 44 | |
| }, | |
| { | |
| "Batch Mean": 0.23638486862182617, | |
| "accuracy": 0.7109375, | |
| "epoch": 0.11, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1125, | |
| "grad_norm": 6.963595390319824, | |
| "learning_rate": 4.671052631578948e-06, | |
| "loss": 0.5261, | |
| "step": 45 | |
| }, | |
| { | |
| "Batch Mean": 0.07831740379333496, | |
| "accuracy": 0.8125, | |
| "epoch": 0.1125, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.115, | |
| "grad_norm": 5.393492221832275, | |
| "learning_rate": 4.657894736842106e-06, | |
| "loss": 0.405, | |
| "step": 46 | |
| }, | |
| { | |
| "Batch Mean": 0.10195636749267578, | |
| "accuracy": 0.75, | |
| "epoch": 0.115, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.1175, | |
| "grad_norm": 9.352888107299805, | |
| "learning_rate": 4.6447368421052635e-06, | |
| "loss": 0.5726, | |
| "step": 47 | |
| }, | |
| { | |
| "Batch Mean": 0.4460281729698181, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.1175, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 9.130428314208984, | |
| "learning_rate": 4.631578947368421e-06, | |
| "loss": 0.5347, | |
| "step": 48 | |
| }, | |
| { | |
| "Batch Mean": -0.2610776424407959, | |
| "accuracy": 0.796875, | |
| "epoch": 0.12, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1225, | |
| "grad_norm": 6.863037586212158, | |
| "learning_rate": 4.618421052631579e-06, | |
| "loss": 0.4456, | |
| "step": 49 | |
| }, | |
| { | |
| "Batch Mean": -0.19384944438934326, | |
| "accuracy": 0.6328125, | |
| "epoch": 0.1225, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 6.617422103881836, | |
| "learning_rate": 4.605263157894737e-06, | |
| "loss": 0.5905, | |
| "step": 50 | |
| }, | |
| { | |
| "Batch Mean": -0.4514150619506836, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.125, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1275, | |
| "grad_norm": 7.564980506896973, | |
| "learning_rate": 4.592105263157895e-06, | |
| "loss": 0.4885, | |
| "step": 51 | |
| }, | |
| { | |
| "Batch Mean": -0.003515481948852539, | |
| "accuracy": 0.734375, | |
| "epoch": 0.1275, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 5.258878707885742, | |
| "learning_rate": 4.578947368421053e-06, | |
| "loss": 0.5032, | |
| "step": 52 | |
| }, | |
| { | |
| "Batch Mean": -0.10907495021820068, | |
| "accuracy": 0.71875, | |
| "epoch": 0.13, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.1325, | |
| "grad_norm": 5.931639194488525, | |
| "learning_rate": 4.565789473684211e-06, | |
| "loss": 0.542, | |
| "step": 53 | |
| }, | |
| { | |
| "Batch Mean": 0.46248769760131836, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.1325, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.135, | |
| "grad_norm": 7.1796112060546875, | |
| "learning_rate": 4.552631578947369e-06, | |
| "loss": 0.4685, | |
| "step": 54 | |
| }, | |
| { | |
| "Batch Mean": 0.1234641969203949, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.135, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1375, | |
| "grad_norm": 6.273733139038086, | |
| "learning_rate": 4.539473684210527e-06, | |
| "loss": 0.4989, | |
| "step": 55 | |
| }, | |
| { | |
| "Batch Mean": -0.15052831172943115, | |
| "accuracy": 0.78125, | |
| "epoch": 0.1375, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 5.630648612976074, | |
| "learning_rate": 4.526315789473685e-06, | |
| "loss": 0.4518, | |
| "step": 56 | |
| }, | |
| { | |
| "Batch Mean": 0.10203218460083008, | |
| "accuracy": 0.75, | |
| "epoch": 0.14, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1425, | |
| "grad_norm": 5.465209484100342, | |
| "learning_rate": 4.513157894736843e-06, | |
| "loss": 0.4775, | |
| "step": 57 | |
| }, | |
| { | |
| "Batch Mean": -0.2146952748298645, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.1425, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.145, | |
| "grad_norm": 5.516883373260498, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.4672, | |
| "step": 58 | |
| }, | |
| { | |
| "Batch Mean": 0.39426422119140625, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.145, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.1475, | |
| "grad_norm": 6.986379146575928, | |
| "learning_rate": 4.4868421052631584e-06, | |
| "loss": 0.4851, | |
| "step": 59 | |
| }, | |
| { | |
| "Batch Mean": 0.1964409053325653, | |
| "accuracy": 0.75, | |
| "epoch": 0.1475, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 6.413231372833252, | |
| "learning_rate": 4.473684210526316e-06, | |
| "loss": 0.5073, | |
| "step": 60 | |
| }, | |
| { | |
| "Batch Mean": -0.21541327238082886, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.15, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1525, | |
| "grad_norm": 5.495061874389648, | |
| "learning_rate": 4.460526315789474e-06, | |
| "loss": 0.4751, | |
| "step": 61 | |
| }, | |
| { | |
| "Batch Mean": -0.3059917688369751, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.1525, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.155, | |
| "grad_norm": 6.379850387573242, | |
| "learning_rate": 4.447368421052632e-06, | |
| "loss": 0.4679, | |
| "step": 62 | |
| }, | |
| { | |
| "Batch Mean": -0.5860270261764526, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.155, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.1575, | |
| "grad_norm": 8.42182731628418, | |
| "learning_rate": 4.43421052631579e-06, | |
| "loss": 0.4152, | |
| "step": 63 | |
| }, | |
| { | |
| "Batch Mean": 0.14216375350952148, | |
| "accuracy": 0.7578125, | |
| "epoch": 0.1575, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 6.759493827819824, | |
| "learning_rate": 4.4210526315789476e-06, | |
| "loss": 0.4852, | |
| "step": 64 | |
| }, | |
| { | |
| "Batch Mean": 0.3515692353248596, | |
| "accuracy": 0.796875, | |
| "epoch": 0.16, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1625, | |
| "grad_norm": 7.567401885986328, | |
| "learning_rate": 4.407894736842105e-06, | |
| "loss": 0.4449, | |
| "step": 65 | |
| }, | |
| { | |
| "Batch Mean": 0.281324565410614, | |
| "accuracy": 0.765625, | |
| "epoch": 0.1625, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.165, | |
| "grad_norm": 7.912868022918701, | |
| "learning_rate": 4.394736842105263e-06, | |
| "loss": 0.5717, | |
| "step": 66 | |
| }, | |
| { | |
| "Batch Mean": 0.07387387752532959, | |
| "accuracy": 0.796875, | |
| "epoch": 0.165, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.1675, | |
| "grad_norm": 5.737614631652832, | |
| "learning_rate": 4.381578947368421e-06, | |
| "loss": 0.4532, | |
| "step": 67 | |
| }, | |
| { | |
| "Batch Mean": -0.45137548446655273, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.1675, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 9.204909324645996, | |
| "learning_rate": 4.368421052631579e-06, | |
| "loss": 0.501, | |
| "step": 68 | |
| }, | |
| { | |
| "Batch Mean": -0.12238574028015137, | |
| "accuracy": 0.828125, | |
| "epoch": 0.17, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.1725, | |
| "grad_norm": 6.6219162940979, | |
| "learning_rate": 4.3552631578947375e-06, | |
| "loss": 0.416, | |
| "step": 69 | |
| }, | |
| { | |
| "Batch Mean": 0.06974506378173828, | |
| "accuracy": 0.78125, | |
| "epoch": 0.1725, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 5.526142120361328, | |
| "learning_rate": 4.342105263157895e-06, | |
| "loss": 0.4364, | |
| "step": 70 | |
| }, | |
| { | |
| "Batch Mean": 0.3401278257369995, | |
| "accuracy": 0.8359375, | |
| "epoch": 0.175, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1775, | |
| "grad_norm": 6.567529678344727, | |
| "learning_rate": 4.328947368421053e-06, | |
| "loss": 0.4396, | |
| "step": 71 | |
| }, | |
| { | |
| "Batch Mean": -0.12450069189071655, | |
| "accuracy": 0.75, | |
| "epoch": 0.1775, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 6.054138660430908, | |
| "learning_rate": 4.315789473684211e-06, | |
| "loss": 0.4604, | |
| "step": 72 | |
| }, | |
| { | |
| "Batch Mean": 0.0036773681640625, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.18, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1825, | |
| "grad_norm": 6.112196445465088, | |
| "learning_rate": 4.302631578947369e-06, | |
| "loss": 0.4091, | |
| "step": 73 | |
| }, | |
| { | |
| "Batch Mean": -0.3432164192199707, | |
| "accuracy": 0.7890625, | |
| "epoch": 0.1825, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.185, | |
| "grad_norm": 6.806588649749756, | |
| "learning_rate": 4.289473684210527e-06, | |
| "loss": 0.4194, | |
| "step": 74 | |
| }, | |
| { | |
| "Batch Mean": -0.10890483856201172, | |
| "accuracy": 0.671875, | |
| "epoch": 0.185, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 7.000553131103516, | |
| "learning_rate": 4.276315789473684e-06, | |
| "loss": 0.5318, | |
| "step": 75 | |
| }, | |
| { | |
| "Batch Mean": 0.6419280171394348, | |
| "accuracy": 0.7265625, | |
| "epoch": 0.1875, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 8.422764778137207, | |
| "learning_rate": 4.2631578947368425e-06, | |
| "loss": 0.4539, | |
| "step": 76 | |
| }, | |
| { | |
| "Batch Mean": 0.28451067209243774, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.19, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.1925, | |
| "grad_norm": 6.512566566467285, | |
| "learning_rate": 4.25e-06, | |
| "loss": 0.4522, | |
| "step": 77 | |
| }, | |
| { | |
| "Batch Mean": -0.07123541831970215, | |
| "accuracy": 0.8125, | |
| "epoch": 0.1925, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.195, | |
| "grad_norm": 6.979062080383301, | |
| "learning_rate": 4.236842105263158e-06, | |
| "loss": 0.4827, | |
| "step": 78 | |
| }, | |
| { | |
| "Batch Mean": -0.2889130115509033, | |
| "accuracy": 0.71875, | |
| "epoch": 0.195, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.1975, | |
| "grad_norm": 7.523491859436035, | |
| "learning_rate": 4.223684210526316e-06, | |
| "loss": 0.4501, | |
| "step": 79 | |
| }, | |
| { | |
| "Batch Mean": -0.2912619113922119, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.1975, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 6.1573662757873535, | |
| "learning_rate": 4.210526315789474e-06, | |
| "loss": 0.4537, | |
| "step": 80 | |
| }, | |
| { | |
| "Batch Mean": 0.07626897096633911, | |
| "accuracy": 0.84375, | |
| "epoch": 0.2, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2025, | |
| "grad_norm": 5.308783054351807, | |
| "learning_rate": 4.197368421052632e-06, | |
| "loss": 0.3777, | |
| "step": 81 | |
| }, | |
| { | |
| "Batch Mean": 0.44231414794921875, | |
| "accuracy": 0.78125, | |
| "epoch": 0.2025, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.205, | |
| "grad_norm": 7.6725969314575195, | |
| "learning_rate": 4.18421052631579e-06, | |
| "loss": 0.4681, | |
| "step": 82 | |
| }, | |
| { | |
| "Batch Mean": 0.05066095292568207, | |
| "accuracy": 0.7734375, | |
| "epoch": 0.205, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.2075, | |
| "grad_norm": 6.136054039001465, | |
| "learning_rate": 4.171052631578948e-06, | |
| "loss": 0.4111, | |
| "step": 83 | |
| }, | |
| { | |
| "Batch Mean": 0.11537289619445801, | |
| "accuracy": 0.8515625, | |
| "epoch": 0.2075, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 5.31056022644043, | |
| "learning_rate": 4.157894736842106e-06, | |
| "loss": 0.345, | |
| "step": 84 | |
| }, | |
| { | |
| "Batch Mean": 0.05826067924499512, | |
| "accuracy": 0.734375, | |
| "epoch": 0.21, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.2125, | |
| "grad_norm": 6.368407249450684, | |
| "learning_rate": 4.144736842105263e-06, | |
| "loss": 0.4834, | |
| "step": 85 | |
| }, | |
| { | |
| "Batch Mean": -0.05756664276123047, | |
| "accuracy": 0.75, | |
| "epoch": 0.2125, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.215, | |
| "grad_norm": 7.063327312469482, | |
| "learning_rate": 4.1315789473684216e-06, | |
| "loss": 0.5013, | |
| "step": 86 | |
| }, | |
| { | |
| "Batch Mean": -0.08471214771270752, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.215, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.2175, | |
| "grad_norm": 6.508166313171387, | |
| "learning_rate": 4.118421052631579e-06, | |
| "loss": 0.4431, | |
| "step": 87 | |
| }, | |
| { | |
| "Batch Mean": -0.06070905923843384, | |
| "accuracy": 0.703125, | |
| "epoch": 0.2175, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 7.8081583976745605, | |
| "learning_rate": 4.105263157894737e-06, | |
| "loss": 0.5481, | |
| "step": 88 | |
| }, | |
| { | |
| "Batch Mean": -0.15125751495361328, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.22, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.2225, | |
| "grad_norm": 5.806808948516846, | |
| "learning_rate": 4.092105263157895e-06, | |
| "loss": 0.3994, | |
| "step": 89 | |
| }, | |
| { | |
| "Batch Mean": -0.0017851591110229492, | |
| "accuracy": 0.828125, | |
| "epoch": 0.2225, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 5.560466766357422, | |
| "learning_rate": 4.078947368421053e-06, | |
| "loss": 0.4, | |
| "step": 90 | |
| }, | |
| { | |
| "Batch Mean": 0.07748031616210938, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.225, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2275, | |
| "grad_norm": 5.754173278808594, | |
| "learning_rate": 4.065789473684211e-06, | |
| "loss": 0.4019, | |
| "step": 91 | |
| }, | |
| { | |
| "Batch Mean": -0.24073825776576996, | |
| "accuracy": 0.78125, | |
| "epoch": 0.2275, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 6.231720447540283, | |
| "learning_rate": 4.052631578947368e-06, | |
| "loss": 0.4585, | |
| "step": 92 | |
| }, | |
| { | |
| "Batch Mean": -0.45170480012893677, | |
| "accuracy": 0.75, | |
| "epoch": 0.23, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2325, | |
| "grad_norm": 7.313357353210449, | |
| "learning_rate": 4.0394736842105265e-06, | |
| "loss": 0.4762, | |
| "step": 93 | |
| }, | |
| { | |
| "Batch Mean": 0.024792194366455078, | |
| "accuracy": 0.796875, | |
| "epoch": 0.2325, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.235, | |
| "grad_norm": 4.933104515075684, | |
| "learning_rate": 4.026315789473684e-06, | |
| "loss": 0.4138, | |
| "step": 94 | |
| }, | |
| { | |
| "Batch Mean": -0.02751898765563965, | |
| "accuracy": 0.7421875, | |
| "epoch": 0.235, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.2375, | |
| "grad_norm": 5.39032506942749, | |
| "learning_rate": 4.013157894736842e-06, | |
| "loss": 0.4787, | |
| "step": 95 | |
| }, | |
| { | |
| "Batch Mean": 0.06753873825073242, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.2375, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 5.1791462898254395, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.4488, | |
| "step": 96 | |
| }, | |
| { | |
| "Batch Mean": 0.22616100311279297, | |
| "accuracy": 0.828125, | |
| "epoch": 0.24, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2425, | |
| "grad_norm": 5.300809383392334, | |
| "learning_rate": 3.986842105263158e-06, | |
| "loss": 0.4142, | |
| "step": 97 | |
| }, | |
| { | |
| "Batch Mean": -0.15787070989608765, | |
| "accuracy": 0.8515625, | |
| "epoch": 0.2425, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 0.245, | |
| "grad_norm": 5.704384803771973, | |
| "learning_rate": 3.9736842105263165e-06, | |
| "loss": 0.3797, | |
| "step": 98 | |
| }, | |
| { | |
| "Batch Mean": 0.13187718391418457, | |
| "accuracy": 0.8046875, | |
| "epoch": 0.245, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.2475, | |
| "grad_norm": 6.045483589172363, | |
| "learning_rate": 3.960526315789474e-06, | |
| "loss": 0.5195, | |
| "step": 99 | |
| }, | |
| { | |
| "Batch Mean": -0.08533608913421631, | |
| "accuracy": 0.8203125, | |
| "epoch": 0.2475, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 4.997817516326904, | |
| "learning_rate": 3.947368421052632e-06, | |
| "loss": 0.3816, | |
| "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 | |
| } | |