| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0977734753146176, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.030977734753146177, | |
| "grad_norm": 99.0, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.9661, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.061955469506292354, | |
| "grad_norm": 374.0, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.2124, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09293320425943853, | |
| "grad_norm": 1392640.0, | |
| "learning_rate": 3e-06, | |
| "loss": 2.2122, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12391093901258471, | |
| "grad_norm": 78.5, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.9339, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15488867376573087, | |
| "grad_norm": 13500416.0, | |
| "learning_rate": 5e-06, | |
| "loss": 1.8448, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18586640851887706, | |
| "grad_norm": 98.0, | |
| "learning_rate": 6e-06, | |
| "loss": 1.9577, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21684414327202323, | |
| "grad_norm": 11337728.0, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 2.089, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.24782187802516942, | |
| "grad_norm": 20736.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.8571, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2787996127783156, | |
| "grad_norm": 8512.0, | |
| "learning_rate": 9e-06, | |
| "loss": 2.0639, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.30977734753146174, | |
| "grad_norm": 1004.0, | |
| "learning_rate": 1e-05, | |
| "loss": 1.978, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.34075508228460794, | |
| "grad_norm": 470.0, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 1.9845, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3717328170377541, | |
| "grad_norm": 268435456.0, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.9825, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4027105517909003, | |
| "grad_norm": 50593792.0, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 2.0309, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.43368828654404645, | |
| "grad_norm": 942080.0, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 1.804, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.46466602129719264, | |
| "grad_norm": 93.5, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.9188, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.49564375605033884, | |
| "grad_norm": 138.0, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.7755, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.526621490803485, | |
| "grad_norm": 2310144.0, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 1.9404, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5575992255566312, | |
| "grad_norm": 252928.0, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.9115, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5885769603097774, | |
| "grad_norm": 36096.0, | |
| "learning_rate": 1.9e-05, | |
| "loss": 2.0781, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6195546950629235, | |
| "grad_norm": 143.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8338, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6505324298160697, | |
| "grad_norm": 222208.0, | |
| "learning_rate": 2.1e-05, | |
| "loss": 1.9065, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6815101645692159, | |
| "grad_norm": 11599872.0, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.8142, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.712487899322362, | |
| "grad_norm": 3568.0, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 1.7811, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7434656340755083, | |
| "grad_norm": 81.5, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.7512, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7744433688286544, | |
| "grad_norm": 3696.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.8697, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8054211035818006, | |
| "grad_norm": 63744.0, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.7677, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8363988383349468, | |
| "grad_norm": 55.5, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 1.7307, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8673765730880929, | |
| "grad_norm": 202.0, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 1.9074, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8983543078412392, | |
| "grad_norm": 348.0, | |
| "learning_rate": 2.9e-05, | |
| "loss": 1.7037, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9293320425943853, | |
| "grad_norm": 342.0, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5886, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9603097773475314, | |
| "grad_norm": 62.5, | |
| "learning_rate": 3.1e-05, | |
| "loss": 1.6942, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9912875121006777, | |
| "grad_norm": 91.5, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.5326, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.0222652468538238, | |
| "grad_norm": 6560.0, | |
| "learning_rate": 3.3e-05, | |
| "loss": 1.6642, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.05324298160697, | |
| "grad_norm": 51.5, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 1.5624, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.084220716360116, | |
| "grad_norm": 1752.0, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.447, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1151984511132624, | |
| "grad_norm": 48.25, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.6133, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1461761858664086, | |
| "grad_norm": 52.25, | |
| "learning_rate": 3.7e-05, | |
| "loss": 1.4596, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.1771539206195547, | |
| "grad_norm": 12032.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.5244, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.2081316553727008, | |
| "grad_norm": 49.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 1.5972, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.239109390125847, | |
| "grad_norm": 39.25, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2712, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2700871248789931, | |
| "grad_norm": 1120.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 1.4318, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.3010648596321395, | |
| "grad_norm": 28160.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.3211, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.3320425943852856, | |
| "grad_norm": 12845056.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.4051, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3630203291384317, | |
| "grad_norm": 6979584.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.2505, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3939980638915779, | |
| "grad_norm": 27.75, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1342, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.424975798644724, | |
| "grad_norm": 73.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.2342, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4559535333978704, | |
| "grad_norm": 19.75, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.0688, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4869312681510165, | |
| "grad_norm": 604.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.0641, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.5179090029041626, | |
| "grad_norm": 79.5, | |
| "learning_rate": 4.9e-05, | |
| "loss": 1.0869, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.5488867376573088, | |
| "grad_norm": 2144.0, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0356, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5488867376573088, | |
| "eval_loss": 0.21533620357513428, | |
| "eval_runtime": 145.9664, | |
| "eval_samples_per_second": 10.276, | |
| "eval_steps_per_second": 2.569, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.579864472410455, | |
| "grad_norm": 125.0, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 1.0417, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.6108422071636013, | |
| "grad_norm": 17.875, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 1.0518, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.6418199419167472, | |
| "grad_norm": 68.0, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 1.1404, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.6727976766698935, | |
| "grad_norm": 19.5, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.9938, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.7037754114230397, | |
| "grad_norm": 290.0, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.9374, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.7347531461761858, | |
| "grad_norm": 58.5, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.0777, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.7657308809293322, | |
| "grad_norm": 422.0, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 1.059, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.796708615682478, | |
| "grad_norm": 28.625, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.965, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.8276863504356244, | |
| "grad_norm": 43.75, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.9527, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.8586640851887706, | |
| "grad_norm": 12.75, | |
| "learning_rate": 6e-05, | |
| "loss": 0.8296, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8896418199419167, | |
| "grad_norm": 16.0, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.933, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.920619554695063, | |
| "grad_norm": 11.875, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.8117, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.951597289448209, | |
| "grad_norm": 50.0, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.9475, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9825750242013553, | |
| "grad_norm": 143.0, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.8241, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.0135527589545013, | |
| "grad_norm": 108.5, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.7861, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.0445304937076476, | |
| "grad_norm": 13184.0, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.8384, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.075508228460794, | |
| "grad_norm": 1736704.0, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.8896, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.10648596321394, | |
| "grad_norm": 48.75, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.9377, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.1374636979670862, | |
| "grad_norm": 2816.0, | |
| "learning_rate": 6.9e-05, | |
| "loss": 0.8322, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.168441432720232, | |
| "grad_norm": 3.125, | |
| "learning_rate": 7e-05, | |
| "loss": 0.8397, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.1994191674733785, | |
| "grad_norm": 8.6875, | |
| "learning_rate": 7.1e-05, | |
| "loss": 0.9265, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.230396902226525, | |
| "grad_norm": 9.875, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.8141, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.261374636979671, | |
| "grad_norm": 14.8125, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.7629, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.292352371732817, | |
| "grad_norm": 6.4375, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.85, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.323330106485963, | |
| "grad_norm": 3.796875, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.9116, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.3543078412391094, | |
| "grad_norm": 544768.0, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.8437, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.3852855759922553, | |
| "grad_norm": 117760.0, | |
| "learning_rate": 7.7e-05, | |
| "loss": 0.9072, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.4162633107454017, | |
| "grad_norm": 22.75, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.0168, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.447241045498548, | |
| "grad_norm": 2211840.0, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 1.1433, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.478218780251694, | |
| "grad_norm": 41.5, | |
| "learning_rate": 8e-05, | |
| "loss": 0.7485, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.5091965150048403, | |
| "grad_norm": 1810432.0, | |
| "learning_rate": 8.1e-05, | |
| "loss": 0.9516, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.5401742497579862, | |
| "grad_norm": 1081344.0, | |
| "learning_rate": 8.2e-05, | |
| "loss": 1.0742, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.5711519845111326, | |
| "grad_norm": 337641472.0, | |
| "learning_rate": 8.3e-05, | |
| "loss": 1.023, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.602129719264279, | |
| "grad_norm": 14548992.0, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.185, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.633107454017425, | |
| "grad_norm": 4.71875, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.3584, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.664085188770571, | |
| "grad_norm": 618496.0, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.9947, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.695062923523717, | |
| "grad_norm": 22151168.0, | |
| "learning_rate": 8.7e-05, | |
| "loss": 1.0296, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.7260406582768635, | |
| "grad_norm": 77824.0, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.8889, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.75701839303001, | |
| "grad_norm": 2.640625, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 0.733, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.7879961277831558, | |
| "grad_norm": 18.625, | |
| "learning_rate": 9e-05, | |
| "loss": 0.7786, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.818973862536302, | |
| "grad_norm": 9.0625, | |
| "learning_rate": 9.1e-05, | |
| "loss": 0.7221, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.849951597289448, | |
| "grad_norm": 10.4375, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.6316, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.8809293320425944, | |
| "grad_norm": 4.84375, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 0.7015, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.9119070667957407, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 9.4e-05, | |
| "loss": 0.7161, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.9428848015488867, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 9.5e-05, | |
| "loss": 0.7325, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.973862536302033, | |
| "grad_norm": 5.90625, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.6447, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.004840271055179, | |
| "grad_norm": 1.6015625, | |
| "learning_rate": 9.7e-05, | |
| "loss": 0.7079, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.0358180058083253, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 9.8e-05, | |
| "loss": 0.6075, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.0667957405614716, | |
| "grad_norm": 1.9140625, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 0.697, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.0977734753146176, | |
| "grad_norm": 1.140625, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7175, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0977734753146176, | |
| "eval_loss": 0.1426076889038086, | |
| "eval_runtime": 146.0018, | |
| "eval_samples_per_second": 10.274, | |
| "eval_steps_per_second": 2.568, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 9660, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0946451186709955e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |