|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.03968253968253968, |
|
"eval_steps": 500, |
|
"global_step": 1000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 23.495065689086914, |
|
"learning_rate": 4.0000000000000003e-07, |
|
"loss": 8.6232, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 24.87381935119629, |
|
"learning_rate": 9e-07, |
|
"loss": 8.4514, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 22.63846778869629, |
|
"learning_rate": 1.4000000000000001e-06, |
|
"loss": 8.3152, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 20.65507698059082, |
|
"learning_rate": 1.9e-06, |
|
"loss": 7.9732, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 16.147356033325195, |
|
"learning_rate": 2.4000000000000003e-06, |
|
"loss": 7.6879, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 14.252687454223633, |
|
"learning_rate": 2.9e-06, |
|
"loss": 7.4152, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 12.011382102966309, |
|
"learning_rate": 3.4000000000000005e-06, |
|
"loss": 7.2472, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 10.081536293029785, |
|
"learning_rate": 3.9e-06, |
|
"loss": 7.2433, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 9.65556526184082, |
|
"learning_rate": 4.4e-06, |
|
"loss": 7.2298, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 8.830799102783203, |
|
"learning_rate": 4.9000000000000005e-06, |
|
"loss": 7.19, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 9.019427299499512, |
|
"learning_rate": 5.4e-06, |
|
"loss": 7.2362, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 9.13768482208252, |
|
"learning_rate": 5.9e-06, |
|
"loss": 7.2804, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 9.02840805053711, |
|
"learning_rate": 6.4000000000000006e-06, |
|
"loss": 7.2345, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.178731918334961, |
|
"learning_rate": 6.900000000000001e-06, |
|
"loss": 7.3014, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.83882999420166, |
|
"learning_rate": 7.4e-06, |
|
"loss": 7.3875, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.3687162399292, |
|
"learning_rate": 7.9e-06, |
|
"loss": 7.4546, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.09121036529541, |
|
"learning_rate": 8.400000000000001e-06, |
|
"loss": 7.407, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 12.731450080871582, |
|
"learning_rate": 8.9e-06, |
|
"loss": 7.4321, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.601162910461426, |
|
"learning_rate": 9.4e-06, |
|
"loss": 7.4155, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.006213188171387, |
|
"learning_rate": 9.900000000000002e-06, |
|
"loss": 7.3705, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.79794454574585, |
|
"learning_rate": 1.04e-05, |
|
"loss": 7.3759, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.281346321105957, |
|
"learning_rate": 1.09e-05, |
|
"loss": 7.3798, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.605266094207764, |
|
"learning_rate": 1.1400000000000001e-05, |
|
"loss": 7.3878, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.446412563323975, |
|
"learning_rate": 1.19e-05, |
|
"loss": 7.3098, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.6324334144592285, |
|
"learning_rate": 1.24e-05, |
|
"loss": 7.2995, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.330737113952637, |
|
"learning_rate": 1.29e-05, |
|
"loss": 7.1887, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.283708572387695, |
|
"learning_rate": 1.3400000000000002e-05, |
|
"loss": 7.1184, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 6.950069904327393, |
|
"learning_rate": 1.3900000000000002e-05, |
|
"loss": 6.9876, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 6.844958782196045, |
|
"learning_rate": 1.44e-05, |
|
"loss": 6.9319, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.499362468719482, |
|
"learning_rate": 1.49e-05, |
|
"loss": 6.9253, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.45572280883789, |
|
"learning_rate": 1.54e-05, |
|
"loss": 6.7999, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.108625411987305, |
|
"learning_rate": 1.59e-05, |
|
"loss": 6.7667, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 8.131596565246582, |
|
"learning_rate": 1.6400000000000002e-05, |
|
"loss": 6.7599, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.2159247398376465, |
|
"learning_rate": 1.69e-05, |
|
"loss": 6.7379, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.335048675537109, |
|
"learning_rate": 1.74e-05, |
|
"loss": 6.6553, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.361081600189209, |
|
"learning_rate": 1.79e-05, |
|
"loss": 6.6905, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"grad_norm": 7.59442138671875, |
|
"learning_rate": 1.84e-05, |
|
"loss": 6.6356, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.31695032119751, |
|
"learning_rate": 1.8900000000000002e-05, |
|
"loss": 6.6538, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.541380882263184, |
|
"learning_rate": 1.94e-05, |
|
"loss": 6.6873, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 8.056180000305176, |
|
"learning_rate": 1.9900000000000003e-05, |
|
"loss": 6.6332, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.74859094619751, |
|
"learning_rate": 2.04e-05, |
|
"loss": 6.5488, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.700525760650635, |
|
"learning_rate": 2.09e-05, |
|
"loss": 6.611, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.347009181976318, |
|
"learning_rate": 2.1400000000000002e-05, |
|
"loss": 6.6621, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.607337951660156, |
|
"learning_rate": 2.19e-05, |
|
"loss": 6.568, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.816825866699219, |
|
"learning_rate": 2.2400000000000002e-05, |
|
"loss": 6.6293, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.984443664550781, |
|
"learning_rate": 2.29e-05, |
|
"loss": 6.5667, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.995021820068359, |
|
"learning_rate": 2.3400000000000003e-05, |
|
"loss": 6.5062, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.876286029815674, |
|
"learning_rate": 2.39e-05, |
|
"loss": 6.5034, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.953365802764893, |
|
"learning_rate": 2.44e-05, |
|
"loss": 6.459, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.246520519256592, |
|
"learning_rate": 2.4900000000000002e-05, |
|
"loss": 6.4359, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"eval_loss": 6.133457660675049, |
|
"eval_runtime": 422.8798, |
|
"eval_samples_per_second": 50.182, |
|
"eval_steps_per_second": 0.393, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.1657819747924805, |
|
"learning_rate": 2.54e-05, |
|
"loss": 6.4372, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.203813552856445, |
|
"learning_rate": 2.5900000000000003e-05, |
|
"loss": 6.3684, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.829228401184082, |
|
"learning_rate": 2.64e-05, |
|
"loss": 6.3275, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.641738414764404, |
|
"learning_rate": 2.6900000000000003e-05, |
|
"loss": 6.2923, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.824960708618164, |
|
"learning_rate": 2.7400000000000002e-05, |
|
"loss": 6.3685, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.7754364013671875, |
|
"learning_rate": 2.7900000000000004e-05, |
|
"loss": 6.3069, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 7.3608903884887695, |
|
"learning_rate": 2.84e-05, |
|
"loss": 6.2915, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.551064968109131, |
|
"learning_rate": 2.8899999999999998e-05, |
|
"loss": 6.1676, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.588657379150391, |
|
"learning_rate": 2.94e-05, |
|
"loss": 6.2311, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.631275177001953, |
|
"learning_rate": 2.9900000000000002e-05, |
|
"loss": 6.1824, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.2990288734436035, |
|
"learning_rate": 3.04e-05, |
|
"loss": 6.1641, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 6.6439714431762695, |
|
"learning_rate": 3.09e-05, |
|
"loss": 6.0125, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.402431964874268, |
|
"learning_rate": 3.1400000000000004e-05, |
|
"loss": 6.1314, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.883937358856201, |
|
"learning_rate": 3.19e-05, |
|
"loss": 5.9858, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.988973140716553, |
|
"learning_rate": 3.24e-05, |
|
"loss": 5.9886, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.186301231384277, |
|
"learning_rate": 3.29e-05, |
|
"loss": 5.9732, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.2982354164123535, |
|
"learning_rate": 3.3400000000000005e-05, |
|
"loss": 6.0391, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.037730693817139, |
|
"learning_rate": 3.3900000000000004e-05, |
|
"loss": 6.0547, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.280949592590332, |
|
"learning_rate": 3.4399999999999996e-05, |
|
"loss": 5.8734, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.861902236938477, |
|
"learning_rate": 3.49e-05, |
|
"loss": 5.8997, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.4033589363098145, |
|
"learning_rate": 3.54e-05, |
|
"loss": 5.8111, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.376819133758545, |
|
"learning_rate": 3.59e-05, |
|
"loss": 5.9027, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.777069568634033, |
|
"learning_rate": 3.6400000000000004e-05, |
|
"loss": 5.727, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.183162689208984, |
|
"learning_rate": 3.69e-05, |
|
"loss": 5.733, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.658604621887207, |
|
"learning_rate": 3.74e-05, |
|
"loss": 5.7018, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.263491153717041, |
|
"learning_rate": 3.79e-05, |
|
"loss": 5.7315, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.86756706237793, |
|
"learning_rate": 3.8400000000000005e-05, |
|
"loss": 5.6637, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.122657299041748, |
|
"learning_rate": 3.8900000000000004e-05, |
|
"loss": 5.6849, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.6024580001831055, |
|
"learning_rate": 3.94e-05, |
|
"loss": 5.5835, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.619604587554932, |
|
"learning_rate": 3.99e-05, |
|
"loss": 5.7, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.7342352867126465, |
|
"learning_rate": 4.0400000000000006e-05, |
|
"loss": 5.4825, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.796237468719482, |
|
"learning_rate": 4.09e-05, |
|
"loss": 5.6496, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.601135730743408, |
|
"learning_rate": 4.14e-05, |
|
"loss": 5.3526, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.576679229736328, |
|
"learning_rate": 4.19e-05, |
|
"loss": 5.3757, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.774380683898926, |
|
"learning_rate": 4.24e-05, |
|
"loss": 5.5164, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.140038967132568, |
|
"learning_rate": 4.29e-05, |
|
"loss": 5.3124, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 5.971868515014648, |
|
"learning_rate": 4.3400000000000005e-05, |
|
"loss": 5.3387, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 6.431824207305908, |
|
"learning_rate": 4.39e-05, |
|
"loss": 5.4363, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 6.174604892730713, |
|
"learning_rate": 4.44e-05, |
|
"loss": 5.282, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 6.313474178314209, |
|
"learning_rate": 4.49e-05, |
|
"loss": 5.3754, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 6.066281318664551, |
|
"learning_rate": 4.5400000000000006e-05, |
|
"loss": 5.3197, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 7.4433817863464355, |
|
"learning_rate": 4.5900000000000004e-05, |
|
"loss": 5.3277, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 5.884599208831787, |
|
"learning_rate": 4.64e-05, |
|
"loss": 5.0681, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 6.17655611038208, |
|
"learning_rate": 4.69e-05, |
|
"loss": 5.1159, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 7.090456485748291, |
|
"learning_rate": 4.74e-05, |
|
"loss": 5.2264, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 6.508240699768066, |
|
"learning_rate": 4.79e-05, |
|
"loss": 5.2113, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 6.5788445472717285, |
|
"learning_rate": 4.8400000000000004e-05, |
|
"loss": 4.9996, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 5.722221374511719, |
|
"learning_rate": 4.89e-05, |
|
"loss": 4.9688, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 5.813933849334717, |
|
"learning_rate": 4.94e-05, |
|
"loss": 4.8691, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"grad_norm": 6.425748348236084, |
|
"learning_rate": 4.99e-05, |
|
"loss": 4.9883, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"eval_loss": 4.425854682922363, |
|
"eval_runtime": 422.6592, |
|
"eval_samples_per_second": 50.208, |
|
"eval_steps_per_second": 0.393, |
|
"step": 1000 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 50400, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 2, |
|
"save_steps": 500, |
|
"total_flos": 0.0, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|