| { | |
| "best_metric": 0.8676381707191467, | |
| "best_model_checkpoint": "./results_v1/checkpoint-10200", | |
| "epoch": 25.0, | |
| "eval_steps": 500, | |
| "global_step": 10200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24509803921568626, | |
| "grad_norm": 10681.4521484375, | |
| "learning_rate": 0.0004950980392156863, | |
| "loss": 2.5241, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.49019607843137253, | |
| "grad_norm": 6339.50927734375, | |
| "learning_rate": 0.0004901960784313725, | |
| "loss": 2.0561, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7352941176470589, | |
| "grad_norm": 16086.7705078125, | |
| "learning_rate": 0.0004852941176470588, | |
| "loss": 1.9379, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9803921568627451, | |
| "grad_norm": 10268.5185546875, | |
| "learning_rate": 0.00048039215686274513, | |
| "loss": 1.8333, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.6950186491012573, | |
| "eval_runtime": 17.2704, | |
| "eval_samples_per_second": 463.222, | |
| "eval_steps_per_second": 3.648, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 1.2254901960784315, | |
| "grad_norm": 10859.7197265625, | |
| "learning_rate": 0.00047549019607843134, | |
| "loss": 1.7372, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 13147.400390625, | |
| "learning_rate": 0.00047058823529411766, | |
| "loss": 1.6714, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.715686274509804, | |
| "grad_norm": 11227.5166015625, | |
| "learning_rate": 0.0004656862745098039, | |
| "loss": 1.5845, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.9607843137254903, | |
| "grad_norm": 10829.8447265625, | |
| "learning_rate": 0.0004607843137254902, | |
| "loss": 1.5283, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.389205813407898, | |
| "eval_runtime": 16.6762, | |
| "eval_samples_per_second": 479.726, | |
| "eval_steps_per_second": 3.778, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 13556.080078125, | |
| "learning_rate": 0.00045588235294117646, | |
| "loss": 1.4638, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.450980392156863, | |
| "grad_norm": 10844.283203125, | |
| "learning_rate": 0.0004509803921568628, | |
| "loss": 1.4344, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.696078431372549, | |
| "grad_norm": 12495.66796875, | |
| "learning_rate": 0.00044607843137254904, | |
| "loss": 1.3873, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 11954.7265625, | |
| "learning_rate": 0.0004411764705882353, | |
| "loss": 1.3733, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.241082787513733, | |
| "eval_runtime": 17.1795, | |
| "eval_samples_per_second": 465.671, | |
| "eval_steps_per_second": 3.667, | |
| "step": 1224 | |
| }, | |
| { | |
| "epoch": 3.186274509803922, | |
| "grad_norm": 12640.005859375, | |
| "learning_rate": 0.0004362745098039216, | |
| "loss": 1.32, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.431372549019608, | |
| "grad_norm": 14839.830078125, | |
| "learning_rate": 0.0004313725490196079, | |
| "loss": 1.2993, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.6764705882352944, | |
| "grad_norm": 17634.357421875, | |
| "learning_rate": 0.0004264705882352941, | |
| "loss": 1.284, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.9215686274509802, | |
| "grad_norm": 11596.447265625, | |
| "learning_rate": 0.00042156862745098043, | |
| "loss": 1.2622, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.152040958404541, | |
| "eval_runtime": 16.7185, | |
| "eval_samples_per_second": 478.513, | |
| "eval_steps_per_second": 3.768, | |
| "step": 1632 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 10679.4697265625, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 1.2385, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.411764705882353, | |
| "grad_norm": 10487.45703125, | |
| "learning_rate": 0.0004117647058823529, | |
| "loss": 1.2131, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.6568627450980395, | |
| "grad_norm": 16033.8701171875, | |
| "learning_rate": 0.0004068627450980392, | |
| "loss": 1.2052, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.901960784313726, | |
| "grad_norm": 13594.8916015625, | |
| "learning_rate": 0.0004019607843137255, | |
| "loss": 1.1878, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.0944446325302124, | |
| "eval_runtime": 16.7053, | |
| "eval_samples_per_second": 478.89, | |
| "eval_steps_per_second": 3.771, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 5.147058823529412, | |
| "grad_norm": 12540.892578125, | |
| "learning_rate": 0.00039705882352941176, | |
| "loss": 1.1714, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.392156862745098, | |
| "grad_norm": 11532.8291015625, | |
| "learning_rate": 0.000392156862745098, | |
| "loss": 1.1554, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 5.637254901960784, | |
| "grad_norm": 13462.1943359375, | |
| "learning_rate": 0.00038725490196078434, | |
| "loss": 1.1487, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 14919.408203125, | |
| "learning_rate": 0.00038235294117647055, | |
| "loss": 1.1286, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.0514171123504639, | |
| "eval_runtime": 17.2713, | |
| "eval_samples_per_second": 463.195, | |
| "eval_steps_per_second": 3.648, | |
| "step": 2448 | |
| }, | |
| { | |
| "epoch": 6.127450980392156, | |
| "grad_norm": 12428.1318359375, | |
| "learning_rate": 0.0003774509803921569, | |
| "loss": 1.1157, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.372549019607844, | |
| "grad_norm": 12143.3505859375, | |
| "learning_rate": 0.00037254901960784314, | |
| "loss": 1.1092, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 6.617647058823529, | |
| "grad_norm": 11514.5810546875, | |
| "learning_rate": 0.00036764705882352946, | |
| "loss": 1.0906, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 6.862745098039216, | |
| "grad_norm": 18561.646484375, | |
| "learning_rate": 0.00036274509803921567, | |
| "loss": 1.0951, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.0217130184173584, | |
| "eval_runtime": 16.6786, | |
| "eval_samples_per_second": 479.657, | |
| "eval_steps_per_second": 3.777, | |
| "step": 2856 | |
| }, | |
| { | |
| "epoch": 7.107843137254902, | |
| "grad_norm": 16182.7216796875, | |
| "learning_rate": 0.000357843137254902, | |
| "loss": 1.0806, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.352941176470588, | |
| "grad_norm": 15809.748046875, | |
| "learning_rate": 0.00035294117647058826, | |
| "loss": 1.0551, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.598039215686274, | |
| "grad_norm": 11234.1923828125, | |
| "learning_rate": 0.0003480392156862745, | |
| "loss": 1.0648, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 7.8431372549019605, | |
| "grad_norm": 13925.9716796875, | |
| "learning_rate": 0.0003431372549019608, | |
| "loss": 1.0551, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9909818172454834, | |
| "eval_runtime": 16.7369, | |
| "eval_samples_per_second": 477.987, | |
| "eval_steps_per_second": 3.764, | |
| "step": 3264 | |
| }, | |
| { | |
| "epoch": 8.088235294117647, | |
| "grad_norm": 13518.5712890625, | |
| "learning_rate": 0.0003382352941176471, | |
| "loss": 1.0471, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 12719.046875, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 1.0271, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 8.57843137254902, | |
| "grad_norm": 13152.646484375, | |
| "learning_rate": 0.0003284313725490196, | |
| "loss": 1.0161, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 11174.1962890625, | |
| "learning_rate": 0.0003235294117647059, | |
| "loss": 1.0259, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.9708221554756165, | |
| "eval_runtime": 17.2479, | |
| "eval_samples_per_second": 463.825, | |
| "eval_steps_per_second": 3.653, | |
| "step": 3672 | |
| }, | |
| { | |
| "epoch": 9.068627450980392, | |
| "grad_norm": 12128.1474609375, | |
| "learning_rate": 0.0003186274509803921, | |
| "loss": 1.0239, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 9.313725490196079, | |
| "grad_norm": 10996.591796875, | |
| "learning_rate": 0.00031372549019607844, | |
| "loss": 0.9963, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 9.558823529411764, | |
| "grad_norm": 12174.8466796875, | |
| "learning_rate": 0.0003088235294117647, | |
| "loss": 1.006, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 9.803921568627452, | |
| "grad_norm": 11552.1796875, | |
| "learning_rate": 0.00030392156862745097, | |
| "loss": 1.0001, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.9560179114341736, | |
| "eval_runtime": 17.2671, | |
| "eval_samples_per_second": 463.308, | |
| "eval_steps_per_second": 3.649, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 10.049019607843137, | |
| "grad_norm": 14723.912109375, | |
| "learning_rate": 0.00029901960784313724, | |
| "loss": 0.9901, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 10.294117647058824, | |
| "grad_norm": 16030.7060546875, | |
| "learning_rate": 0.00029411764705882356, | |
| "loss": 0.9763, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 10.53921568627451, | |
| "grad_norm": 12146.142578125, | |
| "learning_rate": 0.0002892156862745098, | |
| "loss": 0.9638, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 10.784313725490197, | |
| "grad_norm": 12161.0078125, | |
| "learning_rate": 0.0002843137254901961, | |
| "loss": 0.9891, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.9404354095458984, | |
| "eval_runtime": 16.7426, | |
| "eval_samples_per_second": 477.824, | |
| "eval_steps_per_second": 3.763, | |
| "step": 4488 | |
| }, | |
| { | |
| "epoch": 11.029411764705882, | |
| "grad_norm": 11092.498046875, | |
| "learning_rate": 0.00027941176470588236, | |
| "loss": 0.9653, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.27450980392157, | |
| "grad_norm": 15073.44140625, | |
| "learning_rate": 0.0002745098039215687, | |
| "loss": 0.9608, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 11.519607843137255, | |
| "grad_norm": 12903.4189453125, | |
| "learning_rate": 0.0002696078431372549, | |
| "loss": 0.9572, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 14061.0234375, | |
| "learning_rate": 0.0002647058823529412, | |
| "loss": 0.9523, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.9268321394920349, | |
| "eval_runtime": 16.7172, | |
| "eval_samples_per_second": 478.548, | |
| "eval_steps_per_second": 3.769, | |
| "step": 4896 | |
| }, | |
| { | |
| "epoch": 12.009803921568627, | |
| "grad_norm": 13276.89453125, | |
| "learning_rate": 0.0002598039215686275, | |
| "loss": 0.9492, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 12.254901960784313, | |
| "grad_norm": 11237.9267578125, | |
| "learning_rate": 0.00025490196078431374, | |
| "loss": 0.9417, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 11576.23046875, | |
| "learning_rate": 0.00025, | |
| "loss": 0.9379, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 12.745098039215687, | |
| "grad_norm": 14009.544921875, | |
| "learning_rate": 0.00024509803921568627, | |
| "loss": 0.9349, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 12.990196078431373, | |
| "grad_norm": 12153.9677734375, | |
| "learning_rate": 0.00024019607843137256, | |
| "loss": 0.9329, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.9184120893478394, | |
| "eval_runtime": 17.2276, | |
| "eval_samples_per_second": 464.372, | |
| "eval_steps_per_second": 3.657, | |
| "step": 5304 | |
| }, | |
| { | |
| "epoch": 13.235294117647058, | |
| "grad_norm": 12770.306640625, | |
| "learning_rate": 0.00023529411764705883, | |
| "loss": 0.9219, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 13.480392156862745, | |
| "grad_norm": 12070.76171875, | |
| "learning_rate": 0.0002303921568627451, | |
| "loss": 0.9192, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 13.72549019607843, | |
| "grad_norm": 12918.3779296875, | |
| "learning_rate": 0.0002254901960784314, | |
| "loss": 0.925, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 13.970588235294118, | |
| "grad_norm": 11361.103515625, | |
| "learning_rate": 0.00022058823529411765, | |
| "loss": 0.9109, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.9040451645851135, | |
| "eval_runtime": 16.707, | |
| "eval_samples_per_second": 478.841, | |
| "eval_steps_per_second": 3.771, | |
| "step": 5712 | |
| }, | |
| { | |
| "epoch": 14.215686274509803, | |
| "grad_norm": 14390.365234375, | |
| "learning_rate": 0.00021568627450980395, | |
| "loss": 0.8986, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 14.46078431372549, | |
| "grad_norm": 12093.57421875, | |
| "learning_rate": 0.00021078431372549021, | |
| "loss": 0.9064, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 14.705882352941176, | |
| "grad_norm": 9752.62890625, | |
| "learning_rate": 0.00020588235294117645, | |
| "loss": 0.9165, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 14.950980392156863, | |
| "grad_norm": 16142.40625, | |
| "learning_rate": 0.00020098039215686275, | |
| "loss": 0.9027, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.9005336761474609, | |
| "eval_runtime": 17.261, | |
| "eval_samples_per_second": 463.471, | |
| "eval_steps_per_second": 3.65, | |
| "step": 6120 | |
| }, | |
| { | |
| "epoch": 15.196078431372548, | |
| "grad_norm": 14545.95703125, | |
| "learning_rate": 0.000196078431372549, | |
| "loss": 0.9078, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 15.441176470588236, | |
| "grad_norm": 10779.703125, | |
| "learning_rate": 0.00019117647058823528, | |
| "loss": 0.8906, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 15.686274509803921, | |
| "grad_norm": 11541.22265625, | |
| "learning_rate": 0.00018627450980392157, | |
| "loss": 0.8925, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 15.931372549019608, | |
| "grad_norm": 11552.533203125, | |
| "learning_rate": 0.00018137254901960784, | |
| "loss": 0.8872, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8934208154678345, | |
| "eval_runtime": 16.6878, | |
| "eval_samples_per_second": 479.393, | |
| "eval_steps_per_second": 3.775, | |
| "step": 6528 | |
| }, | |
| { | |
| "epoch": 16.176470588235293, | |
| "grad_norm": 11795.5029296875, | |
| "learning_rate": 0.00017647058823529413, | |
| "loss": 0.8854, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 16.42156862745098, | |
| "grad_norm": 11742.291015625, | |
| "learning_rate": 0.0001715686274509804, | |
| "loss": 0.8853, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 16.666666666666668, | |
| "grad_norm": 11037.279296875, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 0.8805, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 16.91176470588235, | |
| "grad_norm": 13446.0478515625, | |
| "learning_rate": 0.00016176470588235295, | |
| "loss": 0.8754, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.8874623775482178, | |
| "eval_runtime": 16.6727, | |
| "eval_samples_per_second": 479.827, | |
| "eval_steps_per_second": 3.779, | |
| "step": 6936 | |
| }, | |
| { | |
| "epoch": 17.15686274509804, | |
| "grad_norm": 13627.7470703125, | |
| "learning_rate": 0.00015686274509803922, | |
| "loss": 0.8801, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.401960784313726, | |
| "grad_norm": 14940.4482421875, | |
| "learning_rate": 0.00015196078431372549, | |
| "loss": 0.8703, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 17.647058823529413, | |
| "grad_norm": 17162.34765625, | |
| "learning_rate": 0.00014705882352941178, | |
| "loss": 0.8732, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 17.892156862745097, | |
| "grad_norm": 13630.0888671875, | |
| "learning_rate": 0.00014215686274509804, | |
| "loss": 0.872, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.8821218013763428, | |
| "eval_runtime": 17.107, | |
| "eval_samples_per_second": 467.645, | |
| "eval_steps_per_second": 3.683, | |
| "step": 7344 | |
| }, | |
| { | |
| "epoch": 18.137254901960784, | |
| "grad_norm": 12382.7177734375, | |
| "learning_rate": 0.00013725490196078434, | |
| "loss": 0.8609, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 18.38235294117647, | |
| "grad_norm": 12689.1533203125, | |
| "learning_rate": 0.0001323529411764706, | |
| "loss": 0.8643, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 18.627450980392158, | |
| "grad_norm": 13463.681640625, | |
| "learning_rate": 0.00012745098039215687, | |
| "loss": 0.8624, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 18.872549019607842, | |
| "grad_norm": 11256.0576171875, | |
| "learning_rate": 0.00012254901960784314, | |
| "loss": 0.8648, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.878471314907074, | |
| "eval_runtime": 16.6776, | |
| "eval_samples_per_second": 479.684, | |
| "eval_steps_per_second": 3.778, | |
| "step": 7752 | |
| }, | |
| { | |
| "epoch": 19.11764705882353, | |
| "grad_norm": 13279.96875, | |
| "learning_rate": 0.00011764705882352942, | |
| "loss": 0.8581, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 19.362745098039216, | |
| "grad_norm": 12439.5205078125, | |
| "learning_rate": 0.0001127450980392157, | |
| "loss": 0.8624, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 19.607843137254903, | |
| "grad_norm": 11728.5849609375, | |
| "learning_rate": 0.00010784313725490197, | |
| "loss": 0.8517, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 19.852941176470587, | |
| "grad_norm": 11004.234375, | |
| "learning_rate": 0.00010294117647058823, | |
| "loss": 0.8476, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.8746265769004822, | |
| "eval_runtime": 16.7166, | |
| "eval_samples_per_second": 478.566, | |
| "eval_steps_per_second": 3.769, | |
| "step": 8160 | |
| }, | |
| { | |
| "epoch": 20.098039215686274, | |
| "grad_norm": 12775.7421875, | |
| "learning_rate": 9.80392156862745e-05, | |
| "loss": 0.8636, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 20.34313725490196, | |
| "grad_norm": 11157.79296875, | |
| "learning_rate": 9.313725490196079e-05, | |
| "loss": 0.8482, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 20.58823529411765, | |
| "grad_norm": 12324.103515625, | |
| "learning_rate": 8.823529411764706e-05, | |
| "loss": 0.844, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 20.833333333333332, | |
| "grad_norm": 12305.396484375, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.847, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.8715333342552185, | |
| "eval_runtime": 16.7289, | |
| "eval_samples_per_second": 478.213, | |
| "eval_steps_per_second": 3.766, | |
| "step": 8568 | |
| }, | |
| { | |
| "epoch": 21.07843137254902, | |
| "grad_norm": 14880.2978515625, | |
| "learning_rate": 7.843137254901961e-05, | |
| "loss": 0.8532, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 21.323529411764707, | |
| "grad_norm": 11197.8076171875, | |
| "learning_rate": 7.352941176470589e-05, | |
| "loss": 0.8484, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 21.568627450980394, | |
| "grad_norm": 10673.302734375, | |
| "learning_rate": 6.862745098039217e-05, | |
| "loss": 0.8395, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 21.813725490196077, | |
| "grad_norm": 10516.341796875, | |
| "learning_rate": 6.372549019607843e-05, | |
| "loss": 0.8472, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.8717733025550842, | |
| "eval_runtime": 16.6953, | |
| "eval_samples_per_second": 479.176, | |
| "eval_steps_per_second": 3.774, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 22.058823529411764, | |
| "grad_norm": 11651.111328125, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.8404, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 22.30392156862745, | |
| "grad_norm": 13144.53515625, | |
| "learning_rate": 5.392156862745099e-05, | |
| "loss": 0.8362, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 22.54901960784314, | |
| "grad_norm": 11174.5419921875, | |
| "learning_rate": 4.901960784313725e-05, | |
| "loss": 0.8405, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 22.794117647058822, | |
| "grad_norm": 10513.1904296875, | |
| "learning_rate": 4.411764705882353e-05, | |
| "loss": 0.8417, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.8683518171310425, | |
| "eval_runtime": 17.2108, | |
| "eval_samples_per_second": 464.825, | |
| "eval_steps_per_second": 3.66, | |
| "step": 9384 | |
| }, | |
| { | |
| "epoch": 23.03921568627451, | |
| "grad_norm": 13963.318359375, | |
| "learning_rate": 3.9215686274509805e-05, | |
| "loss": 0.8438, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 23.284313725490197, | |
| "grad_norm": 15601.267578125, | |
| "learning_rate": 3.4313725490196084e-05, | |
| "loss": 0.8422, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 12826.8515625, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 0.8304, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 23.774509803921568, | |
| "grad_norm": 13277.5234375, | |
| "learning_rate": 2.4509803921568626e-05, | |
| "loss": 0.8357, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.8688600659370422, | |
| "eval_runtime": 16.7261, | |
| "eval_samples_per_second": 478.293, | |
| "eval_steps_per_second": 3.767, | |
| "step": 9792 | |
| }, | |
| { | |
| "epoch": 24.019607843137255, | |
| "grad_norm": 11718.4677734375, | |
| "learning_rate": 1.9607843137254903e-05, | |
| "loss": 0.8361, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 24.264705882352942, | |
| "grad_norm": 13389.1337890625, | |
| "learning_rate": 1.4705882352941177e-05, | |
| "loss": 0.8371, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 24.509803921568626, | |
| "grad_norm": 10126.4189453125, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.8234, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 24.754901960784313, | |
| "grad_norm": 15717.1884765625, | |
| "learning_rate": 4.901960784313726e-06, | |
| "loss": 0.8405, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 27583.41015625, | |
| "learning_rate": 0.0, | |
| "loss": 0.8359, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.8676381707191467, | |
| "eval_runtime": 16.6913, | |
| "eval_samples_per_second": 479.291, | |
| "eval_steps_per_second": 3.774, | |
| "step": 10200 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 10200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 2, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.40829394550784e+16, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } |