| { | |
| "best_metric": 0.41427454352378845, | |
| "best_model_checkpoint": "mikhail_panzo/ceb_b128_le3_s8000/checkpoint-500", | |
| "epoch": 627.4509803921569, | |
| "eval_steps": 500, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.9215686274509802, | |
| "grad_norm": 2.899599313735962, | |
| "learning_rate": 2.4500000000000003e-05, | |
| "loss": 0.7166, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.8431372549019605, | |
| "grad_norm": 3.100567102432251, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.5261, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 3.0574705600738525, | |
| "learning_rate": 7.45e-05, | |
| "loss": 0.4841, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 15.686274509803921, | |
| "grad_norm": 1.0856355428695679, | |
| "learning_rate": 9.95e-05, | |
| "loss": 0.4634, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 19.607843137254903, | |
| "grad_norm": 3.049023389816284, | |
| "learning_rate": 0.0001245, | |
| "loss": 0.4622, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 23.529411764705884, | |
| "grad_norm": 1.3187408447265625, | |
| "learning_rate": 0.0001495, | |
| "loss": 0.4491, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 27.45098039215686, | |
| "grad_norm": 2.879023551940918, | |
| "learning_rate": 0.00017449999999999999, | |
| "loss": 0.4391, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 31.372549019607842, | |
| "grad_norm": 0.908079981803894, | |
| "learning_rate": 0.00019950000000000002, | |
| "loss": 0.4446, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 35.294117647058826, | |
| "grad_norm": 2.698949098587036, | |
| "learning_rate": 0.0002245, | |
| "loss": 0.4475, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 39.21568627450981, | |
| "grad_norm": 4.266517639160156, | |
| "learning_rate": 0.0002495, | |
| "loss": 0.4379, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 39.21568627450981, | |
| "eval_loss": 0.41427454352378845, | |
| "eval_runtime": 6.5876, | |
| "eval_samples_per_second": 27.324, | |
| "eval_steps_per_second": 3.491, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 43.13725490196079, | |
| "grad_norm": 2.933168649673462, | |
| "learning_rate": 0.0002745, | |
| "loss": 0.4306, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 47.05882352941177, | |
| "grad_norm": 2.7896311283111572, | |
| "learning_rate": 0.0002995, | |
| "loss": 0.4502, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 50.98039215686274, | |
| "grad_norm": 2.2012181282043457, | |
| "learning_rate": 0.00032450000000000003, | |
| "loss": 0.4493, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 54.90196078431372, | |
| "grad_norm": 2.1830649375915527, | |
| "learning_rate": 0.0003495, | |
| "loss": 0.4231, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 58.8235294117647, | |
| "grad_norm": 2.837082862854004, | |
| "learning_rate": 0.0003745, | |
| "loss": 0.4237, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 62.745098039215684, | |
| "grad_norm": 2.12803316116333, | |
| "learning_rate": 0.0003995, | |
| "loss": 0.4527, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 66.66666666666667, | |
| "grad_norm": 7.096790790557861, | |
| "learning_rate": 0.0004245, | |
| "loss": 0.4536, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 70.58823529411765, | |
| "grad_norm": 2.911712884902954, | |
| "learning_rate": 0.00044950000000000003, | |
| "loss": 0.4573, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 74.50980392156863, | |
| "grad_norm": 3.439180374145508, | |
| "learning_rate": 0.0004745, | |
| "loss": 0.4726, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 78.43137254901961, | |
| "grad_norm": 4.071909427642822, | |
| "learning_rate": 0.0004995, | |
| "loss": 0.4492, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 78.43137254901961, | |
| "eval_loss": 0.4376925528049469, | |
| "eval_runtime": 6.6325, | |
| "eval_samples_per_second": 27.139, | |
| "eval_steps_per_second": 3.468, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 82.3529411764706, | |
| "grad_norm": 6.268165111541748, | |
| "learning_rate": 0.0005245, | |
| "loss": 0.4562, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 86.27450980392157, | |
| "grad_norm": 2.190829038619995, | |
| "learning_rate": 0.0005495, | |
| "loss": 0.4473, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 90.19607843137256, | |
| "grad_norm": 6.477546215057373, | |
| "learning_rate": 0.0005745, | |
| "loss": 0.4587, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 94.11764705882354, | |
| "grad_norm": 2.3519718647003174, | |
| "learning_rate": 0.0005995000000000001, | |
| "loss": 0.4677, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 98.03921568627452, | |
| "grad_norm": 3.0339012145996094, | |
| "learning_rate": 0.0006245000000000001, | |
| "loss": 0.5425, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 101.96078431372548, | |
| "grad_norm": 5.318001747131348, | |
| "learning_rate": 0.0006490000000000001, | |
| "loss": 0.9214, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 105.88235294117646, | |
| "grad_norm": 0.10876531153917313, | |
| "learning_rate": 0.000674, | |
| "loss": 1.5183, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 109.80392156862744, | |
| "grad_norm": 1.2959858179092407, | |
| "learning_rate": 0.000699, | |
| "loss": 1.497, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 113.72549019607843, | |
| "grad_norm": 0.14063771069049835, | |
| "learning_rate": 0.000724, | |
| "loss": 1.4423, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 117.6470588235294, | |
| "grad_norm": 0.3507501184940338, | |
| "learning_rate": 0.000749, | |
| "loss": 1.4417, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 117.6470588235294, | |
| "eval_loss": 1.4372303485870361, | |
| "eval_runtime": 6.5947, | |
| "eval_samples_per_second": 27.295, | |
| "eval_steps_per_second": 3.488, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 121.56862745098039, | |
| "grad_norm": 0.041443753987550735, | |
| "learning_rate": 0.0007740000000000001, | |
| "loss": 1.4335, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 125.49019607843137, | |
| "grad_norm": 0.12605535984039307, | |
| "learning_rate": 0.000799, | |
| "loss": 1.434, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 129.41176470588235, | |
| "grad_norm": 0.22777751088142395, | |
| "learning_rate": 0.000824, | |
| "loss": 1.4312, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 133.33333333333334, | |
| "grad_norm": 0.317199170589447, | |
| "learning_rate": 0.000849, | |
| "loss": 1.4323, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 137.2549019607843, | |
| "grad_norm": 0.10237721353769302, | |
| "learning_rate": 0.000874, | |
| "loss": 1.4332, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 141.1764705882353, | |
| "grad_norm": 0.23147189617156982, | |
| "learning_rate": 0.0008990000000000001, | |
| "loss": 1.4313, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 145.09803921568627, | |
| "grad_norm": 0.1493494212627411, | |
| "learning_rate": 0.000924, | |
| "loss": 1.4446, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 149.01960784313727, | |
| "grad_norm": 0.10309738665819168, | |
| "learning_rate": 0.000949, | |
| "loss": 1.4315, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 152.94117647058823, | |
| "grad_norm": 0.1577301323413849, | |
| "learning_rate": 0.000974, | |
| "loss": 1.4305, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 156.86274509803923, | |
| "grad_norm": 0.14740918576717377, | |
| "learning_rate": 0.000999, | |
| "loss": 1.4311, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 156.86274509803923, | |
| "eval_loss": 1.4307715892791748, | |
| "eval_runtime": 6.5817, | |
| "eval_samples_per_second": 27.349, | |
| "eval_steps_per_second": 3.495, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 160.7843137254902, | |
| "grad_norm": 0.04897089675068855, | |
| "learning_rate": 0.000992, | |
| "loss": 1.4303, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 164.7058823529412, | |
| "grad_norm": 0.052775364369153976, | |
| "learning_rate": 0.0009836666666666668, | |
| "loss": 1.4293, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 168.62745098039215, | |
| "grad_norm": 0.07608811557292938, | |
| "learning_rate": 0.0009753333333333334, | |
| "loss": 1.4302, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 172.54901960784315, | |
| "grad_norm": 0.18155619502067566, | |
| "learning_rate": 0.000967, | |
| "loss": 1.4298, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 176.47058823529412, | |
| "grad_norm": 0.23861084878444672, | |
| "learning_rate": 0.0009586666666666667, | |
| "loss": 1.4492, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 180.3921568627451, | |
| "grad_norm": 0.07681425660848618, | |
| "learning_rate": 0.0009503333333333334, | |
| "loss": 1.4352, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 184.31372549019608, | |
| "grad_norm": 0.1355891227722168, | |
| "learning_rate": 0.000942, | |
| "loss": 1.4308, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 188.23529411764707, | |
| "grad_norm": 0.17761075496673584, | |
| "learning_rate": 0.0009336666666666666, | |
| "loss": 1.4678, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 192.15686274509804, | |
| "grad_norm": 0.10840369015932083, | |
| "learning_rate": 0.0009253333333333333, | |
| "loss": 1.4288, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 196.07843137254903, | |
| "grad_norm": 0.15987902879714966, | |
| "learning_rate": 0.0009170000000000001, | |
| "loss": 1.4506, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 196.07843137254903, | |
| "eval_loss": 1.4364666938781738, | |
| "eval_runtime": 6.5311, | |
| "eval_samples_per_second": 27.561, | |
| "eval_steps_per_second": 3.522, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "grad_norm": 0.056504905223846436, | |
| "learning_rate": 0.0009086666666666667, | |
| "loss": 1.4315, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 203.92156862745097, | |
| "grad_norm": 0.07093175500631332, | |
| "learning_rate": 0.0009003333333333334, | |
| "loss": 1.4302, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 207.84313725490196, | |
| "grad_norm": 0.04637598991394043, | |
| "learning_rate": 0.000892, | |
| "loss": 1.4306, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 211.76470588235293, | |
| "grad_norm": 0.062212761491537094, | |
| "learning_rate": 0.0008836666666666667, | |
| "loss": 1.4296, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 215.68627450980392, | |
| "grad_norm": 0.1276828944683075, | |
| "learning_rate": 0.0008753333333333333, | |
| "loss": 1.4433, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 219.6078431372549, | |
| "grad_norm": 0.07146938890218735, | |
| "learning_rate": 0.000867, | |
| "loss": 1.4309, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 223.52941176470588, | |
| "grad_norm": 0.10493505001068115, | |
| "learning_rate": 0.0008586666666666668, | |
| "loss": 1.43, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 227.45098039215685, | |
| "grad_norm": 0.11434385925531387, | |
| "learning_rate": 0.0008503333333333334, | |
| "loss": 1.4276, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 231.37254901960785, | |
| "grad_norm": 0.16153867542743683, | |
| "learning_rate": 0.000842, | |
| "loss": 1.4288, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 235.2941176470588, | |
| "grad_norm": 0.14236119389533997, | |
| "learning_rate": 0.0008336666666666667, | |
| "loss": 1.4286, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 235.2941176470588, | |
| "eval_loss": 1.4312187433242798, | |
| "eval_runtime": 6.6382, | |
| "eval_samples_per_second": 27.116, | |
| "eval_steps_per_second": 3.465, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 239.2156862745098, | |
| "grad_norm": 0.08121176809072495, | |
| "learning_rate": 0.0008253333333333334, | |
| "loss": 1.426, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 243.13725490196077, | |
| "grad_norm": 0.04294486716389656, | |
| "learning_rate": 0.000817, | |
| "loss": 1.4278, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 247.05882352941177, | |
| "grad_norm": 0.09214005619287491, | |
| "learning_rate": 0.0008086666666666666, | |
| "loss": 1.4283, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 250.98039215686273, | |
| "grad_norm": 0.06561414152383804, | |
| "learning_rate": 0.0008003333333333333, | |
| "loss": 1.4286, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 254.90196078431373, | |
| "grad_norm": 0.06078485772013664, | |
| "learning_rate": 0.0007920000000000001, | |
| "loss": 1.4275, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 258.8235294117647, | |
| "grad_norm": 0.15776388347148895, | |
| "learning_rate": 0.0007836666666666667, | |
| "loss": 1.4296, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 262.7450980392157, | |
| "grad_norm": 0.07666509598493576, | |
| "learning_rate": 0.0007753333333333334, | |
| "loss": 1.4264, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 266.6666666666667, | |
| "grad_norm": 0.12861433625221252, | |
| "learning_rate": 0.000767, | |
| "loss": 1.4301, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 270.5882352941176, | |
| "grad_norm": 0.058193549513816833, | |
| "learning_rate": 0.0007586666666666667, | |
| "loss": 1.4288, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 274.5098039215686, | |
| "grad_norm": 0.2083207666873932, | |
| "learning_rate": 0.0007503333333333333, | |
| "loss": 1.4285, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 274.5098039215686, | |
| "eval_loss": 1.4328523874282837, | |
| "eval_runtime": 6.505, | |
| "eval_samples_per_second": 27.671, | |
| "eval_steps_per_second": 3.536, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 278.4313725490196, | |
| "grad_norm": 0.1663893461227417, | |
| "learning_rate": 0.000742, | |
| "loss": 1.4281, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 282.3529411764706, | |
| "grad_norm": 0.034747958183288574, | |
| "learning_rate": 0.0007336666666666668, | |
| "loss": 1.4278, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 286.27450980392155, | |
| "grad_norm": 0.0872250571846962, | |
| "learning_rate": 0.0007253333333333334, | |
| "loss": 1.4258, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 290.19607843137254, | |
| "grad_norm": 0.08757229149341583, | |
| "learning_rate": 0.000717, | |
| "loss": 1.4224, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 294.11764705882354, | |
| "grad_norm": 0.1288028061389923, | |
| "learning_rate": 0.0007086666666666667, | |
| "loss": 1.425, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 298.03921568627453, | |
| "grad_norm": 0.08947344869375229, | |
| "learning_rate": 0.0007003333333333334, | |
| "loss": 1.4219, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 301.96078431372547, | |
| "grad_norm": 0.10822741687297821, | |
| "learning_rate": 0.000692, | |
| "loss": 1.4217, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 305.88235294117646, | |
| "grad_norm": 0.08467403054237366, | |
| "learning_rate": 0.0006836666666666666, | |
| "loss": 1.423, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 309.80392156862746, | |
| "grad_norm": 0.06203661486506462, | |
| "learning_rate": 0.0006753333333333333, | |
| "loss": 1.4229, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 313.72549019607845, | |
| "grad_norm": 0.11116231977939606, | |
| "learning_rate": 0.0006670000000000001, | |
| "loss": 1.421, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 313.72549019607845, | |
| "eval_loss": 1.4406124353408813, | |
| "eval_runtime": 6.7269, | |
| "eval_samples_per_second": 26.758, | |
| "eval_steps_per_second": 3.419, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 317.6470588235294, | |
| "grad_norm": 0.22143186628818512, | |
| "learning_rate": 0.0006586666666666667, | |
| "loss": 1.4416, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 321.5686274509804, | |
| "grad_norm": 0.04169169440865517, | |
| "learning_rate": 0.0006503333333333334, | |
| "loss": 1.4239, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 325.4901960784314, | |
| "grad_norm": 0.09313623607158661, | |
| "learning_rate": 0.000642, | |
| "loss": 1.4219, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 329.4117647058824, | |
| "grad_norm": 0.1355171650648117, | |
| "learning_rate": 0.0006336666666666667, | |
| "loss": 1.422, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 333.3333333333333, | |
| "grad_norm": 0.11701258271932602, | |
| "learning_rate": 0.0006253333333333333, | |
| "loss": 1.4288, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 337.2549019607843, | |
| "grad_norm": 0.20015889406204224, | |
| "learning_rate": 0.000617, | |
| "loss": 1.423, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 341.1764705882353, | |
| "grad_norm": 0.061756107956171036, | |
| "learning_rate": 0.0006086666666666668, | |
| "loss": 1.4219, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 345.0980392156863, | |
| "grad_norm": 0.10823976993560791, | |
| "learning_rate": 0.0006003333333333334, | |
| "loss": 1.4222, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 349.01960784313724, | |
| "grad_norm": 0.13286006450653076, | |
| "learning_rate": 0.000592, | |
| "loss": 1.4292, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 352.94117647058823, | |
| "grad_norm": 0.10281714797019958, | |
| "learning_rate": 0.0005836666666666667, | |
| "loss": 1.4228, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 352.94117647058823, | |
| "eval_loss": 1.433663249015808, | |
| "eval_runtime": 6.7809, | |
| "eval_samples_per_second": 26.545, | |
| "eval_steps_per_second": 3.392, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 356.8627450980392, | |
| "grad_norm": 0.045487239956855774, | |
| "learning_rate": 0.0005753333333333334, | |
| "loss": 1.4237, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 360.7843137254902, | |
| "grad_norm": 0.08421700447797775, | |
| "learning_rate": 0.000567, | |
| "loss": 1.4216, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 364.70588235294116, | |
| "grad_norm": 0.03885773941874504, | |
| "learning_rate": 0.0005586666666666666, | |
| "loss": 1.423, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 368.62745098039215, | |
| "grad_norm": 0.06695468723773956, | |
| "learning_rate": 0.0005503333333333333, | |
| "loss": 1.4227, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 372.54901960784315, | |
| "grad_norm": 0.059906505048274994, | |
| "learning_rate": 0.0005420000000000001, | |
| "loss": 1.4216, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 376.47058823529414, | |
| "grad_norm": 0.13581004738807678, | |
| "learning_rate": 0.0005336666666666667, | |
| "loss": 1.4241, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 380.3921568627451, | |
| "grad_norm": 0.10568855702877045, | |
| "learning_rate": 0.0005253333333333334, | |
| "loss": 1.4219, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 384.3137254901961, | |
| "grad_norm": 0.16822956502437592, | |
| "learning_rate": 0.000517, | |
| "loss": 1.4224, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 388.2352941176471, | |
| "grad_norm": 0.11007729917764664, | |
| "learning_rate": 0.0005086666666666667, | |
| "loss": 1.4222, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 392.15686274509807, | |
| "grad_norm": 0.06487352401018143, | |
| "learning_rate": 0.0005003333333333333, | |
| "loss": 1.4232, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 392.15686274509807, | |
| "eval_loss": 1.4328006505966187, | |
| "eval_runtime": 6.65, | |
| "eval_samples_per_second": 27.068, | |
| "eval_steps_per_second": 3.459, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 396.078431372549, | |
| "grad_norm": 0.06124414503574371, | |
| "learning_rate": 0.000492, | |
| "loss": 1.422, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "grad_norm": 0.15419355034828186, | |
| "learning_rate": 0.0004836666666666667, | |
| "loss": 1.4212, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 403.921568627451, | |
| "grad_norm": 0.032941777259111404, | |
| "learning_rate": 0.00047533333333333336, | |
| "loss": 1.4223, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 407.84313725490193, | |
| "grad_norm": 0.1344718337059021, | |
| "learning_rate": 0.000467, | |
| "loss": 1.4223, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 411.7647058823529, | |
| "grad_norm": 0.08072862774133682, | |
| "learning_rate": 0.0004586666666666667, | |
| "loss": 1.4202, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 415.6862745098039, | |
| "grad_norm": 0.10441293567419052, | |
| "learning_rate": 0.0004503333333333333, | |
| "loss": 1.4219, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 419.6078431372549, | |
| "grad_norm": 0.17147202789783478, | |
| "learning_rate": 0.000442, | |
| "loss": 1.4218, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 423.52941176470586, | |
| "grad_norm": 0.0913294106721878, | |
| "learning_rate": 0.0004336666666666667, | |
| "loss": 1.4205, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 427.45098039215685, | |
| "grad_norm": 0.08024486899375916, | |
| "learning_rate": 0.00042533333333333334, | |
| "loss": 1.42, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 431.37254901960785, | |
| "grad_norm": 0.20594315230846405, | |
| "learning_rate": 0.000417, | |
| "loss": 1.4208, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 431.37254901960785, | |
| "eval_loss": 1.4330966472625732, | |
| "eval_runtime": 6.5131, | |
| "eval_samples_per_second": 27.637, | |
| "eval_steps_per_second": 3.531, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 435.29411764705884, | |
| "grad_norm": 0.16393162310123444, | |
| "learning_rate": 0.00040866666666666666, | |
| "loss": 1.4215, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 439.2156862745098, | |
| "grad_norm": 0.12962013483047485, | |
| "learning_rate": 0.0004003333333333333, | |
| "loss": 1.4221, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 443.1372549019608, | |
| "grad_norm": 0.09659314155578613, | |
| "learning_rate": 0.00039200000000000004, | |
| "loss": 1.4197, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 447.05882352941177, | |
| "grad_norm": 0.05516098812222481, | |
| "learning_rate": 0.00038366666666666665, | |
| "loss": 1.4214, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 450.98039215686276, | |
| "grad_norm": 0.08098474889993668, | |
| "learning_rate": 0.00037533333333333337, | |
| "loss": 1.4224, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 454.9019607843137, | |
| "grad_norm": 0.20328585803508759, | |
| "learning_rate": 0.000367, | |
| "loss": 1.4215, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 458.8235294117647, | |
| "grad_norm": 0.04339490830898285, | |
| "learning_rate": 0.0003586666666666667, | |
| "loss": 1.4214, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 462.7450980392157, | |
| "grad_norm": 0.0780976340174675, | |
| "learning_rate": 0.00035033333333333336, | |
| "loss": 1.4215, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 466.6666666666667, | |
| "grad_norm": 0.09538757055997849, | |
| "learning_rate": 0.000342, | |
| "loss": 1.4206, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 470.5882352941176, | |
| "grad_norm": 0.10228724032640457, | |
| "learning_rate": 0.0003336666666666667, | |
| "loss": 1.4211, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 470.5882352941176, | |
| "eval_loss": 1.4311357736587524, | |
| "eval_runtime": 6.8021, | |
| "eval_samples_per_second": 26.462, | |
| "eval_steps_per_second": 3.381, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 474.5098039215686, | |
| "grad_norm": 0.05785190686583519, | |
| "learning_rate": 0.0003253333333333333, | |
| "loss": 1.4195, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 478.4313725490196, | |
| "grad_norm": 0.040976058691740036, | |
| "learning_rate": 0.000317, | |
| "loss": 1.4218, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 482.3529411764706, | |
| "grad_norm": 0.0426504947245121, | |
| "learning_rate": 0.00030866666666666667, | |
| "loss": 1.4218, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 486.27450980392155, | |
| "grad_norm": 0.03735409677028656, | |
| "learning_rate": 0.00030033333333333333, | |
| "loss": 1.4196, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 490.19607843137254, | |
| "grad_norm": 0.12306160479784012, | |
| "learning_rate": 0.000292, | |
| "loss": 1.4222, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 494.11764705882354, | |
| "grad_norm": 0.028285862877964973, | |
| "learning_rate": 0.00028366666666666666, | |
| "loss": 1.4212, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 498.03921568627453, | |
| "grad_norm": 0.13171572983264923, | |
| "learning_rate": 0.0002753333333333333, | |
| "loss": 1.4211, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 501.96078431372547, | |
| "grad_norm": 0.2669251561164856, | |
| "learning_rate": 0.00026700000000000004, | |
| "loss": 1.4206, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 505.88235294117646, | |
| "grad_norm": 0.10635057091712952, | |
| "learning_rate": 0.00025866666666666665, | |
| "loss": 1.4198, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 509.80392156862746, | |
| "grad_norm": 0.04717053100466728, | |
| "learning_rate": 0.00025033333333333336, | |
| "loss": 1.4204, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 509.80392156862746, | |
| "eval_loss": 1.4344011545181274, | |
| "eval_runtime": 6.7898, | |
| "eval_samples_per_second": 26.51, | |
| "eval_steps_per_second": 3.387, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 513.7254901960785, | |
| "grad_norm": 0.028606118634343147, | |
| "learning_rate": 0.000242, | |
| "loss": 1.4203, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 517.6470588235294, | |
| "grad_norm": 0.07845211774110794, | |
| "learning_rate": 0.00023366666666666666, | |
| "loss": 1.42, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 521.5686274509804, | |
| "grad_norm": 0.07225753366947174, | |
| "learning_rate": 0.00022533333333333333, | |
| "loss": 1.4197, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 525.4901960784314, | |
| "grad_norm": 0.038016363978385925, | |
| "learning_rate": 0.00021700000000000002, | |
| "loss": 1.4199, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 529.4117647058823, | |
| "grad_norm": 0.06095300614833832, | |
| "learning_rate": 0.00020866666666666668, | |
| "loss": 1.421, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 533.3333333333334, | |
| "grad_norm": 0.07751613110303879, | |
| "learning_rate": 0.00020033333333333334, | |
| "loss": 1.4212, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 537.2549019607843, | |
| "grad_norm": 0.048244625329971313, | |
| "learning_rate": 0.000192, | |
| "loss": 1.4171, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 541.1764705882352, | |
| "grad_norm": 0.041657689958810806, | |
| "learning_rate": 0.00018366666666666667, | |
| "loss": 1.4206, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 545.0980392156863, | |
| "grad_norm": 0.01942858286201954, | |
| "learning_rate": 0.00017533333333333336, | |
| "loss": 1.4202, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 549.0196078431372, | |
| "grad_norm": 0.0659833624958992, | |
| "learning_rate": 0.00016700000000000002, | |
| "loss": 1.4196, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 549.0196078431372, | |
| "eval_loss": 1.4362238645553589, | |
| "eval_runtime": 6.7359, | |
| "eval_samples_per_second": 26.722, | |
| "eval_steps_per_second": 3.415, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 552.9411764705883, | |
| "grad_norm": 0.06083720177412033, | |
| "learning_rate": 0.00015866666666666668, | |
| "loss": 1.42, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 556.8627450980392, | |
| "grad_norm": 0.07482324540615082, | |
| "learning_rate": 0.00015033333333333335, | |
| "loss": 1.4184, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 560.7843137254902, | |
| "grad_norm": 0.05342550203204155, | |
| "learning_rate": 0.00014199999999999998, | |
| "loss": 1.4186, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 564.7058823529412, | |
| "grad_norm": 0.09053777158260345, | |
| "learning_rate": 0.00013366666666666667, | |
| "loss": 1.4191, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 568.6274509803922, | |
| "grad_norm": 0.08072460442781448, | |
| "learning_rate": 0.00012533333333333334, | |
| "loss": 1.42, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 572.5490196078431, | |
| "grad_norm": 0.09836099296808243, | |
| "learning_rate": 0.00011700000000000001, | |
| "loss": 1.4193, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 576.4705882352941, | |
| "grad_norm": 0.05738500505685806, | |
| "learning_rate": 0.00010866666666666666, | |
| "loss": 1.42, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 580.3921568627451, | |
| "grad_norm": 0.06691340357065201, | |
| "learning_rate": 0.00010033333333333334, | |
| "loss": 1.4185, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 584.3137254901961, | |
| "grad_norm": 0.06185409054160118, | |
| "learning_rate": 9.2e-05, | |
| "loss": 1.4172, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 588.2352941176471, | |
| "grad_norm": 0.14216652512550354, | |
| "learning_rate": 8.366666666666666e-05, | |
| "loss": 1.4194, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 588.2352941176471, | |
| "eval_loss": 1.4361063241958618, | |
| "eval_runtime": 6.733, | |
| "eval_samples_per_second": 26.734, | |
| "eval_steps_per_second": 3.416, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 592.156862745098, | |
| "grad_norm": 0.027007540687918663, | |
| "learning_rate": 7.533333333333334e-05, | |
| "loss": 1.4173, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 596.0784313725491, | |
| "grad_norm": 0.04154467582702637, | |
| "learning_rate": 6.7e-05, | |
| "loss": 1.4186, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 600.0, | |
| "grad_norm": 0.10418181121349335, | |
| "learning_rate": 5.8666666666666665e-05, | |
| "loss": 1.4187, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 603.9215686274509, | |
| "grad_norm": 0.041870325803756714, | |
| "learning_rate": 5.0333333333333335e-05, | |
| "loss": 1.4188, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 607.843137254902, | |
| "grad_norm": 0.08349625766277313, | |
| "learning_rate": 4.2000000000000004e-05, | |
| "loss": 1.4191, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 611.7647058823529, | |
| "grad_norm": 0.08984719216823578, | |
| "learning_rate": 3.366666666666667e-05, | |
| "loss": 1.4174, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 615.6862745098039, | |
| "grad_norm": 0.028009561821818352, | |
| "learning_rate": 2.5333333333333334e-05, | |
| "loss": 1.4203, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 619.6078431372549, | |
| "grad_norm": 0.025170741602778435, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 1.4183, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 623.5294117647059, | |
| "grad_norm": 0.08974526822566986, | |
| "learning_rate": 8.666666666666666e-06, | |
| "loss": 1.4177, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 627.4509803921569, | |
| "grad_norm": 0.023904943838715553, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 1.4184, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 627.4509803921569, | |
| "eval_loss": 1.4349356889724731, | |
| "eval_runtime": 6.7417, | |
| "eval_samples_per_second": 26.7, | |
| "eval_steps_per_second": 3.412, | |
| "step": 8000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 8000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 667, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.7404667537759085e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |