| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 100, | |
| "global_step": 2310, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.021645021645021644, | |
| "grad_norm": 2.59375, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9442, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04329004329004329, | |
| "grad_norm": 1.109375, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6729, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06493506493506493, | |
| "grad_norm": 0.765625, | |
| "learning_rate": 0.00015, | |
| "loss": 0.5235, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08658008658008658, | |
| "grad_norm": 0.87890625, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3883, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10822510822510822, | |
| "grad_norm": 0.8203125, | |
| "learning_rate": 0.00025, | |
| "loss": 0.3269, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12987012987012986, | |
| "grad_norm": 0.5, | |
| "learning_rate": 0.0003, | |
| "loss": 0.2779, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 0.00035, | |
| "loss": 0.2607, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17316017316017315, | |
| "grad_norm": 0.78515625, | |
| "learning_rate": 0.0004, | |
| "loss": 0.246, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19480519480519481, | |
| "grad_norm": 0.474609375, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.2469, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21645021645021645, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 0.0005, | |
| "loss": 0.2327, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21645021645021645, | |
| "eval_loss": 0.1517176628112793, | |
| "eval_runtime": 23.1527, | |
| "eval_samples_per_second": 21.596, | |
| "eval_steps_per_second": 0.691, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.000497737556561086, | |
| "loss": 0.231, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2597402597402597, | |
| "grad_norm": 0.46875, | |
| "learning_rate": 0.000495475113122172, | |
| "loss": 0.214, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2813852813852814, | |
| "grad_norm": 0.4453125, | |
| "learning_rate": 0.000493212669683258, | |
| "loss": 0.2164, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.30303030303030304, | |
| "grad_norm": 0.470703125, | |
| "learning_rate": 0.0004909502262443439, | |
| "loss": 0.2327, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3246753246753247, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.0004886877828054299, | |
| "loss": 0.2104, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3463203463203463, | |
| "grad_norm": 0.4765625, | |
| "learning_rate": 0.00048642533936651587, | |
| "loss": 0.2249, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.36796536796536794, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.0004841628959276018, | |
| "loss": 0.2179, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.38961038961038963, | |
| "grad_norm": 0.453125, | |
| "learning_rate": 0.0004819004524886878, | |
| "loss": 0.2174, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.41125541125541126, | |
| "grad_norm": 0.5390625, | |
| "learning_rate": 0.0004796380090497738, | |
| "loss": 0.2159, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4329004329004329, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.0004773755656108598, | |
| "loss": 0.2166, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4329004329004329, | |
| "eval_loss": 0.13433273136615753, | |
| "eval_runtime": 18.3553, | |
| "eval_samples_per_second": 27.24, | |
| "eval_steps_per_second": 0.872, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 0.44921875, | |
| "learning_rate": 0.00047511312217194567, | |
| "loss": 0.2232, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.00047285067873303167, | |
| "loss": 0.2034, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.49783549783549785, | |
| "grad_norm": 0.36328125, | |
| "learning_rate": 0.00047058823529411766, | |
| "loss": 0.2222, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5194805194805194, | |
| "grad_norm": 0.68359375, | |
| "learning_rate": 0.00046832579185520365, | |
| "loss": 0.2043, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5411255411255411, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 0.0004660633484162896, | |
| "loss": 0.2054, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5627705627705628, | |
| "grad_norm": 0.38671875, | |
| "learning_rate": 0.0004638009049773756, | |
| "loss": 0.1922, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5844155844155844, | |
| "grad_norm": 0.349609375, | |
| "learning_rate": 0.0004615384615384616, | |
| "loss": 0.196, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6060606060606061, | |
| "grad_norm": 0.37109375, | |
| "learning_rate": 0.0004592760180995475, | |
| "loss": 0.2008, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6277056277056277, | |
| "grad_norm": 0.431640625, | |
| "learning_rate": 0.00045701357466063346, | |
| "loss": 0.1983, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6493506493506493, | |
| "grad_norm": 0.4609375, | |
| "learning_rate": 0.00045475113122171945, | |
| "loss": 0.1964, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6493506493506493, | |
| "eval_loss": 0.12437203526496887, | |
| "eval_runtime": 19.6174, | |
| "eval_samples_per_second": 25.488, | |
| "eval_steps_per_second": 0.816, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.670995670995671, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.00045248868778280545, | |
| "loss": 0.1922, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6926406926406926, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 0.00045022624434389144, | |
| "loss": 0.1912, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 0.435546875, | |
| "learning_rate": 0.0004479638009049774, | |
| "loss": 0.1927, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7359307359307359, | |
| "grad_norm": 0.375, | |
| "learning_rate": 0.0004457013574660634, | |
| "loss": 0.1919, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7575757575757576, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.0004434389140271493, | |
| "loss": 0.1941, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7792207792207793, | |
| "grad_norm": 0.50390625, | |
| "learning_rate": 0.0004411764705882353, | |
| "loss": 0.1939, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8008658008658008, | |
| "grad_norm": 0.9765625, | |
| "learning_rate": 0.00043891402714932125, | |
| "loss": 0.2017, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8225108225108225, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 0.00043665158371040724, | |
| "loss": 0.2463, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8441558441558441, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 0.00043438914027149324, | |
| "loss": 0.2453, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8658008658008658, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00043212669683257923, | |
| "loss": 0.2457, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8658008658008658, | |
| "eval_loss": 0.15143588185310364, | |
| "eval_runtime": 20.4597, | |
| "eval_samples_per_second": 24.438, | |
| "eval_steps_per_second": 0.782, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8874458874458875, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.00042986425339366517, | |
| "loss": 0.228, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.2177734375, | |
| "learning_rate": 0.0004276018099547511, | |
| "loss": 0.2139, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9307359307359307, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.0004253393665158371, | |
| "loss": 0.2224, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.375, | |
| "learning_rate": 0.0004230769230769231, | |
| "loss": 0.2108, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.974025974025974, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 0.00042081447963800904, | |
| "loss": 0.2003, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9956709956709957, | |
| "grad_norm": 0.47265625, | |
| "learning_rate": 0.00041855203619909503, | |
| "loss": 0.1827, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.0173160173160174, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 0.000416289592760181, | |
| "loss": 0.1791, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0389610389610389, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 0.00041402714932126697, | |
| "loss": 0.1836, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.0606060606060606, | |
| "grad_norm": 0.36328125, | |
| "learning_rate": 0.0004117647058823529, | |
| "loss": 0.1804, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.0822510822510822, | |
| "grad_norm": 0.375, | |
| "learning_rate": 0.0004095022624434389, | |
| "loss": 0.1977, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0822510822510822, | |
| "eval_loss": 0.1261492371559143, | |
| "eval_runtime": 20.0605, | |
| "eval_samples_per_second": 24.925, | |
| "eval_steps_per_second": 0.798, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.103896103896104, | |
| "grad_norm": 32.5, | |
| "learning_rate": 0.0004072398190045249, | |
| "loss": 0.7592, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.1255411255411256, | |
| "grad_norm": 1.0703125, | |
| "learning_rate": 0.0004049773755656109, | |
| "loss": 0.614, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.1471861471861473, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 0.0004027149321266968, | |
| "loss": 0.2444, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.1688311688311688, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 0.0004004524886877828, | |
| "loss": 0.1821, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "grad_norm": 0.5546875, | |
| "learning_rate": 0.00039819004524886876, | |
| "loss": 0.1811, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.2121212121212122, | |
| "grad_norm": 0.373046875, | |
| "learning_rate": 0.00039592760180995475, | |
| "loss": 0.1717, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.2337662337662338, | |
| "grad_norm": 0.58984375, | |
| "learning_rate": 0.00039366515837104075, | |
| "loss": 0.1725, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.2554112554112553, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.0003914027149321267, | |
| "loss": 0.1713, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.277056277056277, | |
| "grad_norm": 0.58203125, | |
| "learning_rate": 0.0003891402714932127, | |
| "loss": 0.1655, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.2987012987012987, | |
| "grad_norm": 0.8046875, | |
| "learning_rate": 0.0003868778280542987, | |
| "loss": 0.1835, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2987012987012987, | |
| "eval_loss": 0.12451652437448502, | |
| "eval_runtime": 20.7534, | |
| "eval_samples_per_second": 24.092, | |
| "eval_steps_per_second": 0.771, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3203463203463204, | |
| "grad_norm": 0.421875, | |
| "learning_rate": 0.00038461538461538467, | |
| "loss": 0.1626, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.341991341991342, | |
| "grad_norm": 0.439453125, | |
| "learning_rate": 0.00038235294117647055, | |
| "loss": 0.1713, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.00038009049773755655, | |
| "loss": 0.161, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.3852813852813852, | |
| "grad_norm": 0.478515625, | |
| "learning_rate": 0.00037782805429864254, | |
| "loss": 0.1823, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.406926406926407, | |
| "grad_norm": 0.404296875, | |
| "learning_rate": 0.00037556561085972854, | |
| "loss": 0.1822, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.4296875, | |
| "learning_rate": 0.0003733031674208145, | |
| "loss": 0.1675, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.4502164502164503, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.00037104072398190047, | |
| "loss": 0.1736, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.4718614718614718, | |
| "grad_norm": 0.412109375, | |
| "learning_rate": 0.00036877828054298646, | |
| "loss": 0.1614, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.4935064935064934, | |
| "grad_norm": 0.373046875, | |
| "learning_rate": 0.0003665158371040724, | |
| "loss": 0.1795, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.5151515151515151, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.00036425339366515834, | |
| "loss": 0.183, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5151515151515151, | |
| "eval_loss": 0.10722990334033966, | |
| "eval_runtime": 19.3953, | |
| "eval_samples_per_second": 25.779, | |
| "eval_steps_per_second": 0.825, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5367965367965368, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.00036199095022624434, | |
| "loss": 0.1647, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.5584415584415585, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.00035972850678733033, | |
| "loss": 0.1649, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.5800865800865802, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.0003574660633484163, | |
| "loss": 0.1676, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.601731601731602, | |
| "grad_norm": 0.41796875, | |
| "learning_rate": 0.00035520361990950226, | |
| "loss": 0.1625, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.6233766233766234, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.00035294117647058826, | |
| "loss": 0.1714, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.645021645021645, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.0003506787330316742, | |
| "loss": 0.1612, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 0.0003484162895927602, | |
| "loss": 0.1623, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.6883116883116882, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.00034615384615384613, | |
| "loss": 0.1508, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.70995670995671, | |
| "grad_norm": 0.34375, | |
| "learning_rate": 0.0003438914027149321, | |
| "loss": 0.1632, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.7316017316017316, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 0.0003416289592760181, | |
| "loss": 0.1682, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7316017316017316, | |
| "eval_loss": 0.10919010639190674, | |
| "eval_runtime": 27.8108, | |
| "eval_samples_per_second": 17.979, | |
| "eval_steps_per_second": 0.575, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7532467532467533, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.0003393665158371041, | |
| "loss": 0.1514, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.774891774891775, | |
| "grad_norm": 0.359375, | |
| "learning_rate": 0.00033710407239819005, | |
| "loss": 0.1536, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.7965367965367967, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 0.000334841628959276, | |
| "loss": 0.1723, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.484375, | |
| "learning_rate": 0.000332579185520362, | |
| "loss": 0.168, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.8398268398268398, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.000330316742081448, | |
| "loss": 0.1515, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.8614718614718615, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.0003280542986425339, | |
| "loss": 0.1694, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.883116883116883, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 0.0003257918552036199, | |
| "loss": 0.1679, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.0003235294117647059, | |
| "loss": 0.1632, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.9264069264069263, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.0003212669683257919, | |
| "loss": 0.1619, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.948051948051948, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.0003190045248868778, | |
| "loss": 0.1632, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.948051948051948, | |
| "eval_loss": 0.11044134944677353, | |
| "eval_runtime": 18.793, | |
| "eval_samples_per_second": 26.606, | |
| "eval_steps_per_second": 0.851, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9696969696969697, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.0003167420814479638, | |
| "loss": 0.1625, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.9913419913419914, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 0.0003144796380090498, | |
| "loss": 0.1616, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.012987012987013, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.00031221719457013577, | |
| "loss": 0.1433, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.034632034632035, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.0003099547511312217, | |
| "loss": 0.1483, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.0562770562770565, | |
| "grad_norm": 0.322265625, | |
| "learning_rate": 0.0003076923076923077, | |
| "loss": 0.1398, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.0779220779220777, | |
| "grad_norm": 0.83203125, | |
| "learning_rate": 0.0003054298642533937, | |
| "loss": 0.1488, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.0995670995670994, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.00030316742081447964, | |
| "loss": 0.1468, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.121212121212121, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.0003009049773755656, | |
| "loss": 0.1503, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 0.00029864253393665157, | |
| "loss": 0.1326, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 0.00029638009049773756, | |
| "loss": 0.1335, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "eval_loss": 0.11199549585580826, | |
| "eval_runtime": 20.4528, | |
| "eval_samples_per_second": 24.447, | |
| "eval_steps_per_second": 0.782, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.186147186147186, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.00029411764705882356, | |
| "loss": 0.1375, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.207792207792208, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.00029185520361990955, | |
| "loss": 0.1334, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.2294372294372296, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.0002895927601809955, | |
| "loss": 0.1332, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.2510822510822512, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.00028733031674208143, | |
| "loss": 0.1327, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.0002850678733031674, | |
| "loss": 0.141, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.2943722943722946, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.0002828054298642534, | |
| "loss": 0.1432, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.316017316017316, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.00028054298642533936, | |
| "loss": 0.1447, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.3376623376623376, | |
| "grad_norm": 0.2890625, | |
| "learning_rate": 0.00027828054298642535, | |
| "loss": 0.1376, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.3593073593073592, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.00027601809954751135, | |
| "loss": 0.1287, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.00027375565610859734, | |
| "loss": 0.1399, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "eval_loss": 0.09553591907024384, | |
| "eval_runtime": 20.0139, | |
| "eval_samples_per_second": 24.983, | |
| "eval_steps_per_second": 0.799, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.4025974025974026, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 0.0002714932126696832, | |
| "loss": 0.1417, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.4242424242424243, | |
| "grad_norm": 0.294921875, | |
| "learning_rate": 0.0002692307692307692, | |
| "loss": 0.1428, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.445887445887446, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.0002669683257918552, | |
| "loss": 0.1367, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.4675324675324677, | |
| "grad_norm": 0.271484375, | |
| "learning_rate": 0.0002647058823529412, | |
| "loss": 0.1383, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.4891774891774894, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00026244343891402715, | |
| "loss": 0.1438, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.5108225108225106, | |
| "grad_norm": 0.3515625, | |
| "learning_rate": 0.00026018099547511314, | |
| "loss": 0.1394, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.5324675324675323, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.00025791855203619913, | |
| "loss": 0.1358, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.554112554112554, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0002556561085972851, | |
| "loss": 0.1266, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.5757575757575757, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.000253393665158371, | |
| "loss": 0.1353, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.000251131221719457, | |
| "loss": 0.1483, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "eval_loss": 0.10388709604740143, | |
| "eval_runtime": 19.843, | |
| "eval_samples_per_second": 25.198, | |
| "eval_steps_per_second": 0.806, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.619047619047619, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 0.000248868778280543, | |
| "loss": 0.1364, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.6406926406926408, | |
| "grad_norm": 0.2080078125, | |
| "learning_rate": 0.000246606334841629, | |
| "loss": 0.1285, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.6623376623376624, | |
| "grad_norm": 0.251953125, | |
| "learning_rate": 0.00024434389140271494, | |
| "loss": 0.1264, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.683982683982684, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.0002420814479638009, | |
| "loss": 0.1331, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.7056277056277054, | |
| "grad_norm": 0.396484375, | |
| "learning_rate": 0.0002398190045248869, | |
| "loss": 0.15, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.00023755656108597284, | |
| "loss": 0.1367, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.7489177489177488, | |
| "grad_norm": 0.275390625, | |
| "learning_rate": 0.00023529411764705883, | |
| "loss": 0.13, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.7705627705627704, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.0002330316742081448, | |
| "loss": 0.1324, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.792207792207792, | |
| "grad_norm": 0.302734375, | |
| "learning_rate": 0.0002307692307692308, | |
| "loss": 0.1336, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.813852813852814, | |
| "grad_norm": 0.23828125, | |
| "learning_rate": 0.00022850678733031673, | |
| "loss": 0.1314, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.813852813852814, | |
| "eval_loss": 0.10149528831243515, | |
| "eval_runtime": 19.1293, | |
| "eval_samples_per_second": 26.138, | |
| "eval_steps_per_second": 0.836, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.8354978354978355, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.00022624434389140272, | |
| "loss": 0.1309, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.0002239819004524887, | |
| "loss": 0.1353, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.878787878787879, | |
| "grad_norm": 0.462890625, | |
| "learning_rate": 0.00022171945701357466, | |
| "loss": 0.1446, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.9004329004329006, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 0.00021945701357466062, | |
| "loss": 0.1333, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.9220779220779223, | |
| "grad_norm": 0.2080078125, | |
| "learning_rate": 0.00021719457013574662, | |
| "loss": 0.1383, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.9437229437229435, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 0.00021493212669683259, | |
| "loss": 0.1496, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.965367965367965, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.00021266968325791855, | |
| "loss": 0.1399, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.987012987012987, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.00021040723981900452, | |
| "loss": 0.139, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.0086580086580086, | |
| "grad_norm": 0.201171875, | |
| "learning_rate": 0.0002081447963800905, | |
| "loss": 0.127, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.0303030303030303, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 0.00020588235294117645, | |
| "loss": 0.114, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.0303030303030303, | |
| "eval_loss": 0.09961362928152084, | |
| "eval_runtime": 20.1151, | |
| "eval_samples_per_second": 24.857, | |
| "eval_steps_per_second": 0.795, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.051948051948052, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.00020361990950226245, | |
| "loss": 0.1222, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.0735930735930737, | |
| "grad_norm": 0.2490234375, | |
| "learning_rate": 0.0002013574660633484, | |
| "loss": 0.1174, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.0952380952380953, | |
| "grad_norm": 0.2275390625, | |
| "learning_rate": 0.00019909502262443438, | |
| "loss": 0.121, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 3.116883116883117, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.00019683257918552037, | |
| "loss": 0.1179, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 3.1385281385281387, | |
| "grad_norm": 0.2373046875, | |
| "learning_rate": 0.00019457013574660634, | |
| "loss": 0.1163, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.16017316017316, | |
| "grad_norm": 0.232421875, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 0.1173, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 3.1818181818181817, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 0.00019004524886877827, | |
| "loss": 0.1193, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 3.2034632034632033, | |
| "grad_norm": 0.458984375, | |
| "learning_rate": 0.00018778280542986427, | |
| "loss": 0.1208, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 3.225108225108225, | |
| "grad_norm": 0.279296875, | |
| "learning_rate": 0.00018552036199095024, | |
| "loss": 0.1229, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 3.2467532467532467, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 0.0001832579185520362, | |
| "loss": 0.1202, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2467532467532467, | |
| "eval_loss": 0.10085491091012955, | |
| "eval_runtime": 21.5392, | |
| "eval_samples_per_second": 23.214, | |
| "eval_steps_per_second": 0.743, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2683982683982684, | |
| "grad_norm": 0.1943359375, | |
| "learning_rate": 0.00018099547511312217, | |
| "loss": 0.119, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 3.29004329004329, | |
| "grad_norm": 0.45703125, | |
| "learning_rate": 0.00017873303167420816, | |
| "loss": 0.124, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 3.311688311688312, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 0.00017647058823529413, | |
| "loss": 0.1214, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0001742081447963801, | |
| "loss": 0.1236, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 3.354978354978355, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.00017194570135746606, | |
| "loss": 0.1175, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.3766233766233764, | |
| "grad_norm": 0.2236328125, | |
| "learning_rate": 0.00016968325791855206, | |
| "loss": 0.1163, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.398268398268398, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.000167420814479638, | |
| "loss": 0.1114, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 3.41991341991342, | |
| "grad_norm": 0.1845703125, | |
| "learning_rate": 0.000165158371040724, | |
| "loss": 0.1164, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.4415584415584415, | |
| "grad_norm": 0.2060546875, | |
| "learning_rate": 0.00016289592760180996, | |
| "loss": 0.1128, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 3.463203463203463, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 0.00016063348416289595, | |
| "loss": 0.1163, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.463203463203463, | |
| "eval_loss": 0.10191706568002701, | |
| "eval_runtime": 23.6847, | |
| "eval_samples_per_second": 21.111, | |
| "eval_steps_per_second": 0.676, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.484848484848485, | |
| "grad_norm": 0.287109375, | |
| "learning_rate": 0.0001583710407239819, | |
| "loss": 0.1171, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 3.5064935064935066, | |
| "grad_norm": 0.310546875, | |
| "learning_rate": 0.00015610859728506788, | |
| "loss": 0.1159, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 3.5281385281385282, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 0.12, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 3.54978354978355, | |
| "grad_norm": 0.208984375, | |
| "learning_rate": 0.00015158371040723982, | |
| "loss": 0.1048, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 0.1591796875, | |
| "learning_rate": 0.00014932126696832579, | |
| "loss": 0.1061, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.5930735930735933, | |
| "grad_norm": 0.416015625, | |
| "learning_rate": 0.00014705882352941178, | |
| "loss": 0.1283, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 3.6147186147186146, | |
| "grad_norm": 0.248046875, | |
| "learning_rate": 0.00014479638009049775, | |
| "loss": 0.1194, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.0001425339366515837, | |
| "loss": 0.1132, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 3.658008658008658, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.00014027149321266968, | |
| "loss": 0.1226, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 3.6796536796536796, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00013800904977375567, | |
| "loss": 0.1245, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.6796536796536796, | |
| "eval_loss": 0.09824839979410172, | |
| "eval_runtime": 19.2671, | |
| "eval_samples_per_second": 25.951, | |
| "eval_steps_per_second": 0.83, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.7012987012987013, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.0001357466063348416, | |
| "loss": 0.1219, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 3.722943722943723, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.0001334841628959276, | |
| "loss": 0.1184, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 3.7445887445887447, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.00013122171945701357, | |
| "loss": 0.1157, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 3.7662337662337664, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 0.00012895927601809957, | |
| "loss": 0.1075, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 3.787878787878788, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 0.0001266968325791855, | |
| "loss": 0.1102, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 0.28125, | |
| "learning_rate": 0.0001244343891402715, | |
| "loss": 0.1184, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 3.8311688311688314, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.00012217194570135747, | |
| "loss": 0.1081, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 3.8528138528138527, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.00011990950226244345, | |
| "loss": 0.1222, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.8744588744588744, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.00011764705882352942, | |
| "loss": 0.1224, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 3.896103896103896, | |
| "grad_norm": 0.228515625, | |
| "learning_rate": 0.0001153846153846154, | |
| "loss": 0.1032, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.896103896103896, | |
| "eval_loss": 0.09573096036911011, | |
| "eval_runtime": 23.7902, | |
| "eval_samples_per_second": 21.017, | |
| "eval_steps_per_second": 0.673, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.9177489177489178, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.00011312217194570136, | |
| "loss": 0.1182, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 3.9393939393939394, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.00011085972850678733, | |
| "loss": 0.1191, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 3.961038961038961, | |
| "grad_norm": 0.3359375, | |
| "learning_rate": 0.00010859728506787331, | |
| "loss": 0.1125, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 3.982683982683983, | |
| "grad_norm": 0.2138671875, | |
| "learning_rate": 0.00010633484162895928, | |
| "loss": 0.1188, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 4.004329004329004, | |
| "grad_norm": 0.30078125, | |
| "learning_rate": 0.00010407239819004526, | |
| "loss": 0.1203, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.025974025974026, | |
| "grad_norm": 0.232421875, | |
| "learning_rate": 0.00010180995475113122, | |
| "loss": 0.1077, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 4.0476190476190474, | |
| "grad_norm": 0.2080078125, | |
| "learning_rate": 9.954751131221719e-05, | |
| "loss": 0.1065, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 4.06926406926407, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 9.728506787330317e-05, | |
| "loss": 0.1112, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 4.090909090909091, | |
| "grad_norm": 0.2333984375, | |
| "learning_rate": 9.502262443438914e-05, | |
| "loss": 0.1006, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 4.112554112554113, | |
| "grad_norm": 0.205078125, | |
| "learning_rate": 9.276018099547512e-05, | |
| "loss": 0.1043, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.112554112554113, | |
| "eval_loss": 0.09408007562160492, | |
| "eval_runtime": 19.9413, | |
| "eval_samples_per_second": 25.074, | |
| "eval_steps_per_second": 0.802, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.134199134199134, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 9.049773755656108e-05, | |
| "loss": 0.1013, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 4.1558441558441555, | |
| "grad_norm": 0.2109375, | |
| "learning_rate": 8.823529411764706e-05, | |
| "loss": 0.1058, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 4.177489177489178, | |
| "grad_norm": 0.19140625, | |
| "learning_rate": 8.597285067873303e-05, | |
| "loss": 0.1069, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 4.199134199134199, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 8.3710407239819e-05, | |
| "loss": 0.1089, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 4.220779220779221, | |
| "grad_norm": 0.283203125, | |
| "learning_rate": 8.144796380090498e-05, | |
| "loss": 0.1026, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.242424242424242, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 7.918552036199095e-05, | |
| "loss": 0.1043, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 4.264069264069264, | |
| "grad_norm": 0.21484375, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.1136, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 7.466063348416289e-05, | |
| "loss": 0.1045, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 4.307359307359308, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 7.239819004524887e-05, | |
| "loss": 0.1032, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "grad_norm": 0.19140625, | |
| "learning_rate": 7.013574660633484e-05, | |
| "loss": 0.0977, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "eval_loss": 0.09685727208852768, | |
| "eval_runtime": 20.4475, | |
| "eval_samples_per_second": 24.453, | |
| "eval_steps_per_second": 0.782, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.35064935064935, | |
| "grad_norm": 0.2333984375, | |
| "learning_rate": 6.78733031674208e-05, | |
| "loss": 0.1019, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 4.372294372294372, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 6.561085972850679e-05, | |
| "loss": 0.1076, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 4.393939393939394, | |
| "grad_norm": 0.1650390625, | |
| "learning_rate": 6.334841628959275e-05, | |
| "loss": 0.0983, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 4.415584415584416, | |
| "grad_norm": 0.21484375, | |
| "learning_rate": 6.108597285067873e-05, | |
| "loss": 0.1023, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 4.437229437229437, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 0.0982, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 4.458874458874459, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 5.656108597285068e-05, | |
| "loss": 0.1071, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 4.48051948051948, | |
| "grad_norm": 0.2265625, | |
| "learning_rate": 5.4298642533936655e-05, | |
| "loss": 0.1095, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 4.5021645021645025, | |
| "grad_norm": 0.240234375, | |
| "learning_rate": 5.203619909502263e-05, | |
| "loss": 0.1105, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 4.523809523809524, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 4.9773755656108595e-05, | |
| "loss": 0.0961, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 0.2412109375, | |
| "learning_rate": 4.751131221719457e-05, | |
| "loss": 0.1016, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "eval_loss": 0.09792981296777725, | |
| "eval_runtime": 20.8232, | |
| "eval_samples_per_second": 24.012, | |
| "eval_steps_per_second": 0.768, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.567099567099567, | |
| "grad_norm": 0.3125, | |
| "learning_rate": 4.524886877828054e-05, | |
| "loss": 0.0953, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 4.588744588744589, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 4.2986425339366516e-05, | |
| "loss": 0.105, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 4.6103896103896105, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 4.072398190045249e-05, | |
| "loss": 0.1063, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 4.632034632034632, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.0982, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 4.653679653679654, | |
| "grad_norm": 0.265625, | |
| "learning_rate": 3.6199095022624436e-05, | |
| "loss": 0.1068, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.675324675324675, | |
| "grad_norm": 0.2421875, | |
| "learning_rate": 3.39366515837104e-05, | |
| "loss": 0.1111, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 4.696969696969697, | |
| "grad_norm": 0.189453125, | |
| "learning_rate": 3.167420814479638e-05, | |
| "loss": 0.1084, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 4.7186147186147185, | |
| "grad_norm": 0.2734375, | |
| "learning_rate": 2.9411764705882354e-05, | |
| "loss": 0.103, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 4.740259740259741, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 2.7149321266968327e-05, | |
| "loss": 0.1089, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 2.4886877828054298e-05, | |
| "loss": 0.097, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "eval_loss": 0.09769493341445923, | |
| "eval_runtime": 19.6594, | |
| "eval_samples_per_second": 25.433, | |
| "eval_steps_per_second": 0.814, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.783549783549784, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 2.262443438914027e-05, | |
| "loss": 0.1083, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 4.805194805194805, | |
| "grad_norm": 0.2158203125, | |
| "learning_rate": 2.0361990950226245e-05, | |
| "loss": 0.1046, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 4.8268398268398265, | |
| "grad_norm": 0.26953125, | |
| "learning_rate": 1.8099547511312218e-05, | |
| "loss": 0.1139, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 4.848484848484849, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 1.583710407239819e-05, | |
| "loss": 0.1038, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 4.87012987012987, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 1.3574660633484164e-05, | |
| "loss": 0.1089, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.891774891774892, | |
| "grad_norm": 0.23046875, | |
| "learning_rate": 1.1312217194570136e-05, | |
| "loss": 0.0922, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 4.913419913419913, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 9.049773755656109e-06, | |
| "loss": 0.1041, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 4.935064935064935, | |
| "grad_norm": 0.255859375, | |
| "learning_rate": 6.787330316742082e-06, | |
| "loss": 0.0933, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 4.956709956709957, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 4.5248868778280546e-06, | |
| "loss": 0.1089, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 4.978354978354979, | |
| "grad_norm": 0.296875, | |
| "learning_rate": 2.2624434389140273e-06, | |
| "loss": 0.1094, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.978354978354979, | |
| "eval_loss": 0.09713861346244812, | |
| "eval_runtime": 20.5975, | |
| "eval_samples_per_second": 24.275, | |
| "eval_steps_per_second": 0.777, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.3359375, | |
| "learning_rate": 0.0, | |
| "loss": 0.1107, | |
| "step": 2310 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2310, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 5.556491100242671e+18, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |