| { | |
| "best_global_step": null, | |
| "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": 5.5244035720825195, | |
| "learning_rate": 1.8e-05, | |
| "loss": 5.7094, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04329004329004329, | |
| "grad_norm": 3.6613221168518066, | |
| "learning_rate": 3.8e-05, | |
| "loss": 5.4625, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06493506493506493, | |
| "grad_norm": 1.8148497343063354, | |
| "learning_rate": 5.8e-05, | |
| "loss": 5.3, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08658008658008658, | |
| "grad_norm": 1.9286500215530396, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 5.2125, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10822510822510822, | |
| "grad_norm": 2.1979825496673584, | |
| "learning_rate": 9.8e-05, | |
| "loss": 5.0969, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12987012987012986, | |
| "grad_norm": 2.6625876426696777, | |
| "learning_rate": 0.000118, | |
| "loss": 4.9938, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 2.470909595489502, | |
| "learning_rate": 0.000138, | |
| "loss": 4.9344, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17316017316017315, | |
| "grad_norm": 2.0229310989379883, | |
| "learning_rate": 0.00015800000000000002, | |
| "loss": 4.7344, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19480519480519481, | |
| "grad_norm": 3.0014655590057373, | |
| "learning_rate": 0.00017800000000000002, | |
| "loss": 4.675, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21645021645021645, | |
| "grad_norm": 3.8314504623413086, | |
| "learning_rate": 0.00019800000000000002, | |
| "loss": 4.6344, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21645021645021645, | |
| "eval_loss": 3.2876250743865967, | |
| "eval_runtime": 6.4678, | |
| "eval_samples_per_second": 77.306, | |
| "eval_steps_per_second": 1.237, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": 2.9059081077575684, | |
| "learning_rate": 0.00019918552036199096, | |
| "loss": 4.475, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2597402597402597, | |
| "grad_norm": 3.183814764022827, | |
| "learning_rate": 0.00019828054298642536, | |
| "loss": 4.4594, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2813852813852814, | |
| "grad_norm": 3.528064012527466, | |
| "learning_rate": 0.00019737556561085976, | |
| "loss": 4.4094, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.30303030303030304, | |
| "grad_norm": 2.935842752456665, | |
| "learning_rate": 0.00019647058823529413, | |
| "loss": 4.2938, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3246753246753247, | |
| "grad_norm": 2.8355188369750977, | |
| "learning_rate": 0.0001955656108597285, | |
| "loss": 4.3031, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3463203463203463, | |
| "grad_norm": 2.983397960662842, | |
| "learning_rate": 0.0001946606334841629, | |
| "loss": 4.2656, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.36796536796536794, | |
| "grad_norm": 3.0755467414855957, | |
| "learning_rate": 0.0001937556561085973, | |
| "loss": 4.1625, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.38961038961038963, | |
| "grad_norm": 2.8793647289276123, | |
| "learning_rate": 0.0001928506787330317, | |
| "loss": 4.1688, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.41125541125541126, | |
| "grad_norm": 2.5316121578216553, | |
| "learning_rate": 0.0001919457013574661, | |
| "loss": 4.1688, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4329004329004329, | |
| "grad_norm": 4.25211763381958, | |
| "learning_rate": 0.00019104072398190046, | |
| "loss": 4.1969, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4329004329004329, | |
| "eval_loss": 2.902250051498413, | |
| "eval_runtime": 6.5111, | |
| "eval_samples_per_second": 76.792, | |
| "eval_steps_per_second": 1.229, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 2.9996044635772705, | |
| "learning_rate": 0.00019013574660633483, | |
| "loss": 4.0812, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 3.2504937648773193, | |
| "learning_rate": 0.00018923076923076923, | |
| "loss": 4.0406, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.49783549783549785, | |
| "grad_norm": 3.9455931186676025, | |
| "learning_rate": 0.00018832579185520362, | |
| "loss": 4.0187, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5194805194805194, | |
| "grad_norm": 4.604526996612549, | |
| "learning_rate": 0.00018742081447963802, | |
| "loss": 4.0031, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5411255411255411, | |
| "grad_norm": 4.142082691192627, | |
| "learning_rate": 0.00018651583710407242, | |
| "loss": 4.0219, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5627705627705628, | |
| "grad_norm": 4.169658660888672, | |
| "learning_rate": 0.0001856108597285068, | |
| "loss": 3.9609, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5844155844155844, | |
| "grad_norm": 3.3557050228118896, | |
| "learning_rate": 0.0001847058823529412, | |
| "loss": 4.0078, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6060606060606061, | |
| "grad_norm": 2.9046578407287598, | |
| "learning_rate": 0.00018380090497737556, | |
| "loss": 3.9531, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6277056277056277, | |
| "grad_norm": 3.487697124481201, | |
| "learning_rate": 0.00018289592760180996, | |
| "loss": 3.9312, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6493506493506493, | |
| "grad_norm": 3.1762382984161377, | |
| "learning_rate": 0.00018199095022624435, | |
| "loss": 3.9469, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6493506493506493, | |
| "eval_loss": 2.724375009536743, | |
| "eval_runtime": 6.3055, | |
| "eval_samples_per_second": 79.295, | |
| "eval_steps_per_second": 1.269, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.670995670995671, | |
| "grad_norm": 3.686875581741333, | |
| "learning_rate": 0.00018108597285067875, | |
| "loss": 3.9625, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6926406926406926, | |
| "grad_norm": 3.1810693740844727, | |
| "learning_rate": 0.00018018099547511315, | |
| "loss": 3.9109, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 3.2442595958709717, | |
| "learning_rate": 0.00017927601809954752, | |
| "loss": 3.8609, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7359307359307359, | |
| "grad_norm": 3.7046546936035156, | |
| "learning_rate": 0.00017837104072398192, | |
| "loss": 3.9125, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7575757575757576, | |
| "grad_norm": 2.9039487838745117, | |
| "learning_rate": 0.0001774660633484163, | |
| "loss": 3.9406, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7792207792207793, | |
| "grad_norm": 3.293367862701416, | |
| "learning_rate": 0.00017656108597285068, | |
| "loss": 3.8234, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8008658008658008, | |
| "grad_norm": 3.5505335330963135, | |
| "learning_rate": 0.00017565610859728508, | |
| "loss": 3.8047, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8225108225108225, | |
| "grad_norm": 3.220608949661255, | |
| "learning_rate": 0.00017475113122171948, | |
| "loss": 3.8516, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8441558441558441, | |
| "grad_norm": 3.3645002841949463, | |
| "learning_rate": 0.00017384615384615385, | |
| "loss": 3.8766, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8658008658008658, | |
| "grad_norm": 3.1517069339752197, | |
| "learning_rate": 0.00017294117647058825, | |
| "loss": 3.8516, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8658008658008658, | |
| "eval_loss": 2.6535000801086426, | |
| "eval_runtime": 6.3052, | |
| "eval_samples_per_second": 79.299, | |
| "eval_steps_per_second": 1.269, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8874458874458875, | |
| "grad_norm": 3.14215350151062, | |
| "learning_rate": 0.00017203619909502264, | |
| "loss": 3.7781, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 3.1268439292907715, | |
| "learning_rate": 0.00017113122171945702, | |
| "loss": 3.8203, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9307359307359307, | |
| "grad_norm": 3.415076494216919, | |
| "learning_rate": 0.0001702262443438914, | |
| "loss": 3.7625, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 3.4878830909729004, | |
| "learning_rate": 0.0001693212669683258, | |
| "loss": 3.8531, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.974025974025974, | |
| "grad_norm": 3.972776174545288, | |
| "learning_rate": 0.00016841628959276018, | |
| "loss": 3.8266, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9956709956709957, | |
| "grad_norm": 4.266445636749268, | |
| "learning_rate": 0.00016751131221719458, | |
| "loss": 3.7516, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.0173160173160174, | |
| "grad_norm": 4.685347080230713, | |
| "learning_rate": 0.00016660633484162898, | |
| "loss": 3.6875, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0389610389610389, | |
| "grad_norm": 3.38425612449646, | |
| "learning_rate": 0.00016570135746606337, | |
| "loss": 3.6719, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.0606060606060606, | |
| "grad_norm": 3.108266592025757, | |
| "learning_rate": 0.00016479638009049774, | |
| "loss": 3.7156, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.0822510822510822, | |
| "grad_norm": 3.597536563873291, | |
| "learning_rate": 0.00016389140271493214, | |
| "loss": 3.6531, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0822510822510822, | |
| "eval_loss": 2.546750068664551, | |
| "eval_runtime": 6.5359, | |
| "eval_samples_per_second": 76.501, | |
| "eval_steps_per_second": 1.224, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.103896103896104, | |
| "grad_norm": 3.6433794498443604, | |
| "learning_rate": 0.0001629864253393665, | |
| "loss": 3.6594, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.1255411255411256, | |
| "grad_norm": 3.4697842597961426, | |
| "learning_rate": 0.0001620814479638009, | |
| "loss": 3.6609, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.1471861471861473, | |
| "grad_norm": 3.0815842151641846, | |
| "learning_rate": 0.0001611764705882353, | |
| "loss": 3.6578, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.1688311688311688, | |
| "grad_norm": 3.8230438232421875, | |
| "learning_rate": 0.0001602714932126697, | |
| "loss": 3.6953, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "grad_norm": 3.399728298187256, | |
| "learning_rate": 0.00015936651583710408, | |
| "loss": 3.5969, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.2121212121212122, | |
| "grad_norm": 3.2932376861572266, | |
| "learning_rate": 0.00015846153846153847, | |
| "loss": 3.65, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.2337662337662338, | |
| "grad_norm": 4.999845504760742, | |
| "learning_rate": 0.00015755656108597284, | |
| "loss": 3.5953, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.2554112554112553, | |
| "grad_norm": 4.074374675750732, | |
| "learning_rate": 0.00015665158371040724, | |
| "loss": 3.6891, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.277056277056277, | |
| "grad_norm": 4.196053504943848, | |
| "learning_rate": 0.00015574660633484164, | |
| "loss": 3.5922, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.2987012987012987, | |
| "grad_norm": 4.6053853034973145, | |
| "learning_rate": 0.00015484162895927604, | |
| "loss": 3.6859, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2987012987012987, | |
| "eval_loss": 2.4678750038146973, | |
| "eval_runtime": 6.7301, | |
| "eval_samples_per_second": 74.293, | |
| "eval_steps_per_second": 1.189, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3203463203463204, | |
| "grad_norm": 4.082015037536621, | |
| "learning_rate": 0.00015393665158371043, | |
| "loss": 3.6109, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.341991341991342, | |
| "grad_norm": 3.0748109817504883, | |
| "learning_rate": 0.0001530316742081448, | |
| "loss": 3.5875, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 3.769073486328125, | |
| "learning_rate": 0.00015212669683257917, | |
| "loss": 3.6469, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.3852813852813852, | |
| "grad_norm": 3.995649576187134, | |
| "learning_rate": 0.00015122171945701357, | |
| "loss": 3.5516, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.406926406926407, | |
| "grad_norm": 3.429675579071045, | |
| "learning_rate": 0.00015031674208144797, | |
| "loss": 3.6437, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 3.3972747325897217, | |
| "learning_rate": 0.00014941176470588237, | |
| "loss": 3.5578, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.4502164502164503, | |
| "grad_norm": 4.749250888824463, | |
| "learning_rate": 0.00014850678733031676, | |
| "loss": 3.5703, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.4718614718614718, | |
| "grad_norm": 5.253241062164307, | |
| "learning_rate": 0.00014760180995475113, | |
| "loss": 3.5125, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.4935064935064934, | |
| "grad_norm": 4.124709129333496, | |
| "learning_rate": 0.0001466968325791855, | |
| "loss": 3.5766, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.5151515151515151, | |
| "grad_norm": 4.052198886871338, | |
| "learning_rate": 0.0001457918552036199, | |
| "loss": 3.6094, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5151515151515151, | |
| "eval_loss": 2.42787504196167, | |
| "eval_runtime": 6.6116, | |
| "eval_samples_per_second": 75.625, | |
| "eval_steps_per_second": 1.21, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5367965367965368, | |
| "grad_norm": 5.0124616622924805, | |
| "learning_rate": 0.0001448868778280543, | |
| "loss": 3.4797, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.5584415584415585, | |
| "grad_norm": 4.067927360534668, | |
| "learning_rate": 0.0001439819004524887, | |
| "loss": 3.5406, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.5800865800865802, | |
| "grad_norm": 3.851562023162842, | |
| "learning_rate": 0.0001430769230769231, | |
| "loss": 3.5719, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.601731601731602, | |
| "grad_norm": 3.5033013820648193, | |
| "learning_rate": 0.00014217194570135747, | |
| "loss": 3.5531, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.6233766233766234, | |
| "grad_norm": 4.524363040924072, | |
| "learning_rate": 0.00014126696832579186, | |
| "loss": 3.5406, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.645021645021645, | |
| "grad_norm": 3.595323085784912, | |
| "learning_rate": 0.00014036199095022623, | |
| "loss": 3.525, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 5.544644832611084, | |
| "learning_rate": 0.00013945701357466063, | |
| "loss": 3.5438, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.6883116883116882, | |
| "grad_norm": 3.835862636566162, | |
| "learning_rate": 0.00013855203619909503, | |
| "loss": 3.5312, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.70995670995671, | |
| "grad_norm": 3.518122911453247, | |
| "learning_rate": 0.00013764705882352943, | |
| "loss": 3.5516, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.7316017316017316, | |
| "grad_norm": 6.213132381439209, | |
| "learning_rate": 0.0001367420814479638, | |
| "loss": 3.5328, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7316017316017316, | |
| "eval_loss": 2.3873751163482666, | |
| "eval_runtime": 6.5472, | |
| "eval_samples_per_second": 76.369, | |
| "eval_steps_per_second": 1.222, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7532467532467533, | |
| "grad_norm": 4.263932704925537, | |
| "learning_rate": 0.0001358371040723982, | |
| "loss": 3.5094, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.774891774891775, | |
| "grad_norm": 3.5482687950134277, | |
| "learning_rate": 0.0001349321266968326, | |
| "loss": 3.55, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.7965367965367967, | |
| "grad_norm": 4.2857346534729, | |
| "learning_rate": 0.00013402714932126696, | |
| "loss": 3.4828, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 3.9592478275299072, | |
| "learning_rate": 0.00013312217194570136, | |
| "loss": 3.5328, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.8398268398268398, | |
| "grad_norm": 3.499769449234009, | |
| "learning_rate": 0.00013221719457013576, | |
| "loss": 3.5, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.8614718614718615, | |
| "grad_norm": 3.270787477493286, | |
| "learning_rate": 0.00013131221719457013, | |
| "loss": 3.4516, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.883116883116883, | |
| "grad_norm": 4.917684078216553, | |
| "learning_rate": 0.00013040723981900453, | |
| "loss": 3.4656, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 4.024767875671387, | |
| "learning_rate": 0.00012950226244343892, | |
| "loss": 3.4922, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.9264069264069263, | |
| "grad_norm": 4.4510650634765625, | |
| "learning_rate": 0.00012859728506787332, | |
| "loss": 3.4891, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.948051948051948, | |
| "grad_norm": 3.7398886680603027, | |
| "learning_rate": 0.0001276923076923077, | |
| "loss": 3.4438, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.948051948051948, | |
| "eval_loss": 2.3561251163482666, | |
| "eval_runtime": 6.5958, | |
| "eval_samples_per_second": 75.806, | |
| "eval_steps_per_second": 1.213, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9696969696969697, | |
| "grad_norm": 4.9417219161987305, | |
| "learning_rate": 0.0001267873303167421, | |
| "loss": 3.4953, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.9913419913419914, | |
| "grad_norm": 4.173361301422119, | |
| "learning_rate": 0.0001258823529411765, | |
| "loss": 3.4813, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.012987012987013, | |
| "grad_norm": 3.6130735874176025, | |
| "learning_rate": 0.00012497737556561086, | |
| "loss": 3.4188, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.034632034632035, | |
| "grad_norm": 4.494176864624023, | |
| "learning_rate": 0.00012407239819004525, | |
| "loss": 3.3375, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.0562770562770565, | |
| "grad_norm": 4.0216593742370605, | |
| "learning_rate": 0.00012316742081447965, | |
| "loss": 3.3531, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.0779220779220777, | |
| "grad_norm": 4.224644184112549, | |
| "learning_rate": 0.00012226244343891405, | |
| "loss": 3.3859, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.0995670995670994, | |
| "grad_norm": 3.7151124477386475, | |
| "learning_rate": 0.0001213574660633484, | |
| "loss": 3.4297, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.121212121212121, | |
| "grad_norm": 4.0024847984313965, | |
| "learning_rate": 0.0001204524886877828, | |
| "loss": 3.3578, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 4.19533109664917, | |
| "learning_rate": 0.0001195475113122172, | |
| "loss": 3.4, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 4.282599925994873, | |
| "learning_rate": 0.00011864253393665159, | |
| "loss": 3.4547, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "eval_loss": 2.300874948501587, | |
| "eval_runtime": 6.3309, | |
| "eval_samples_per_second": 78.978, | |
| "eval_steps_per_second": 1.264, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.186147186147186, | |
| "grad_norm": 3.4820311069488525, | |
| "learning_rate": 0.00011773755656108598, | |
| "loss": 3.3906, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.207792207792208, | |
| "grad_norm": 3.9142370223999023, | |
| "learning_rate": 0.00011683257918552038, | |
| "loss": 3.3203, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.2294372294372296, | |
| "grad_norm": 4.8086628913879395, | |
| "learning_rate": 0.00011592760180995476, | |
| "loss": 3.2969, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.2510822510822512, | |
| "grad_norm": 3.546926975250244, | |
| "learning_rate": 0.00011502262443438914, | |
| "loss": 3.3578, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 4.613463878631592, | |
| "learning_rate": 0.00011411764705882353, | |
| "loss": 3.3609, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.2943722943722946, | |
| "grad_norm": 4.487024784088135, | |
| "learning_rate": 0.00011321266968325792, | |
| "loss": 3.3141, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.316017316017316, | |
| "grad_norm": 3.963923692703247, | |
| "learning_rate": 0.00011230769230769231, | |
| "loss": 3.3312, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.3376623376623376, | |
| "grad_norm": 4.741418361663818, | |
| "learning_rate": 0.00011140271493212671, | |
| "loss": 3.3937, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.3593073593073592, | |
| "grad_norm": 4.762458801269531, | |
| "learning_rate": 0.0001104977375565611, | |
| "loss": 3.3469, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 4.140308380126953, | |
| "learning_rate": 0.0001095927601809955, | |
| "loss": 3.4047, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "eval_loss": 2.2771248817443848, | |
| "eval_runtime": 6.4697, | |
| "eval_samples_per_second": 77.284, | |
| "eval_steps_per_second": 1.237, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.4025974025974026, | |
| "grad_norm": 3.9786176681518555, | |
| "learning_rate": 0.00010868778280542986, | |
| "loss": 3.4047, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.4242424242424243, | |
| "grad_norm": 4.1307501792907715, | |
| "learning_rate": 0.00010778280542986425, | |
| "loss": 3.3984, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.445887445887446, | |
| "grad_norm": 5.001722812652588, | |
| "learning_rate": 0.00010687782805429865, | |
| "loss": 3.3969, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.4675324675324677, | |
| "grad_norm": 4.192739009857178, | |
| "learning_rate": 0.00010597285067873304, | |
| "loss": 3.3328, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.4891774891774894, | |
| "grad_norm": 4.113678932189941, | |
| "learning_rate": 0.00010506787330316743, | |
| "loss": 3.3609, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.5108225108225106, | |
| "grad_norm": 3.677635669708252, | |
| "learning_rate": 0.00010416289592760182, | |
| "loss": 3.3062, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.5324675324675323, | |
| "grad_norm": 4.670088291168213, | |
| "learning_rate": 0.00010325791855203621, | |
| "loss": 3.3344, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.554112554112554, | |
| "grad_norm": 4.220792770385742, | |
| "learning_rate": 0.00010235294117647058, | |
| "loss": 3.2266, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.5757575757575757, | |
| "grad_norm": 4.443659782409668, | |
| "learning_rate": 0.00010144796380090498, | |
| "loss": 3.3875, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "grad_norm": 4.4685187339782715, | |
| "learning_rate": 0.00010054298642533937, | |
| "loss": 3.3391, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "eval_loss": 2.250687599182129, | |
| "eval_runtime": 6.5637, | |
| "eval_samples_per_second": 76.177, | |
| "eval_steps_per_second": 1.219, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.619047619047619, | |
| "grad_norm": 4.526357650756836, | |
| "learning_rate": 9.963800904977376e-05, | |
| "loss": 3.2641, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.6406926406926408, | |
| "grad_norm": 4.371001720428467, | |
| "learning_rate": 9.873303167420816e-05, | |
| "loss": 3.3141, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.6623376623376624, | |
| "grad_norm": 3.9310309886932373, | |
| "learning_rate": 9.782805429864254e-05, | |
| "loss": 3.3391, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.683982683982684, | |
| "grad_norm": 4.313389778137207, | |
| "learning_rate": 9.692307692307692e-05, | |
| "loss": 3.2766, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.7056277056277054, | |
| "grad_norm": 4.895615100860596, | |
| "learning_rate": 9.601809954751132e-05, | |
| "loss": 3.2609, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 3.97171688079834, | |
| "learning_rate": 9.51131221719457e-05, | |
| "loss": 3.2781, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.7489177489177488, | |
| "grad_norm": 4.3905029296875, | |
| "learning_rate": 9.420814479638009e-05, | |
| "loss": 3.2953, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.7705627705627704, | |
| "grad_norm": 4.197689533233643, | |
| "learning_rate": 9.330316742081449e-05, | |
| "loss": 3.2766, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.792207792207792, | |
| "grad_norm": 4.34577751159668, | |
| "learning_rate": 9.239819004524888e-05, | |
| "loss": 3.3141, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.813852813852814, | |
| "grad_norm": 3.8355207443237305, | |
| "learning_rate": 9.149321266968326e-05, | |
| "loss": 3.2859, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.813852813852814, | |
| "eval_loss": 2.2236249446868896, | |
| "eval_runtime": 6.5337, | |
| "eval_samples_per_second": 76.527, | |
| "eval_steps_per_second": 1.224, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.8354978354978355, | |
| "grad_norm": 3.9101665019989014, | |
| "learning_rate": 9.058823529411765e-05, | |
| "loss": 3.2984, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 4.794166564941406, | |
| "learning_rate": 8.968325791855205e-05, | |
| "loss": 3.3188, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.878787878787879, | |
| "grad_norm": 4.448386192321777, | |
| "learning_rate": 8.877828054298642e-05, | |
| "loss": 3.2797, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.9004329004329006, | |
| "grad_norm": 3.9681570529937744, | |
| "learning_rate": 8.787330316742082e-05, | |
| "loss": 3.3484, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.9220779220779223, | |
| "grad_norm": 4.010087966918945, | |
| "learning_rate": 8.696832579185522e-05, | |
| "loss": 3.3297, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.9437229437229435, | |
| "grad_norm": 3.9292078018188477, | |
| "learning_rate": 8.60633484162896e-05, | |
| "loss": 3.3344, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.965367965367965, | |
| "grad_norm": 4.463669776916504, | |
| "learning_rate": 8.515837104072398e-05, | |
| "loss": 3.275, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.987012987012987, | |
| "grad_norm": 3.4622855186462402, | |
| "learning_rate": 8.425339366515838e-05, | |
| "loss": 3.3078, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.0086580086580086, | |
| "grad_norm": 3.9492321014404297, | |
| "learning_rate": 8.334841628959277e-05, | |
| "loss": 3.3141, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.0303030303030303, | |
| "grad_norm": 4.027843475341797, | |
| "learning_rate": 8.244343891402715e-05, | |
| "loss": 3.2484, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.0303030303030303, | |
| "eval_loss": 2.197937488555908, | |
| "eval_runtime": 6.7816, | |
| "eval_samples_per_second": 73.728, | |
| "eval_steps_per_second": 1.18, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.051948051948052, | |
| "grad_norm": 4.027857780456543, | |
| "learning_rate": 8.153846153846155e-05, | |
| "loss": 3.1766, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.0735930735930737, | |
| "grad_norm": 5.006868362426758, | |
| "learning_rate": 8.063348416289593e-05, | |
| "loss": 3.1719, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.0952380952380953, | |
| "grad_norm": 4.768033027648926, | |
| "learning_rate": 7.972850678733033e-05, | |
| "loss": 3.1344, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 3.116883116883117, | |
| "grad_norm": 5.115345478057861, | |
| "learning_rate": 7.882352941176471e-05, | |
| "loss": 3.1719, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 3.1385281385281387, | |
| "grad_norm": 3.7323756217956543, | |
| "learning_rate": 7.79185520361991e-05, | |
| "loss": 3.1344, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.16017316017316, | |
| "grad_norm": 5.077029228210449, | |
| "learning_rate": 7.70135746606335e-05, | |
| "loss": 3.2687, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 3.1818181818181817, | |
| "grad_norm": 4.608593463897705, | |
| "learning_rate": 7.610859728506788e-05, | |
| "loss": 3.2031, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 3.2034632034632033, | |
| "grad_norm": 5.431169033050537, | |
| "learning_rate": 7.520361990950226e-05, | |
| "loss": 3.1984, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 3.225108225108225, | |
| "grad_norm": 4.630691051483154, | |
| "learning_rate": 7.429864253393666e-05, | |
| "loss": 3.2359, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 3.2467532467532467, | |
| "grad_norm": 4.724872589111328, | |
| "learning_rate": 7.339366515837104e-05, | |
| "loss": 3.2109, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2467532467532467, | |
| "eval_loss": 2.1664373874664307, | |
| "eval_runtime": 6.556, | |
| "eval_samples_per_second": 76.266, | |
| "eval_steps_per_second": 1.22, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2683982683982684, | |
| "grad_norm": 4.629438877105713, | |
| "learning_rate": 7.248868778280543e-05, | |
| "loss": 3.1844, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 3.29004329004329, | |
| "grad_norm": 5.131148815155029, | |
| "learning_rate": 7.158371040723983e-05, | |
| "loss": 3.1688, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 3.311688311688312, | |
| "grad_norm": 4.398173809051514, | |
| "learning_rate": 7.067873303167421e-05, | |
| "loss": 3.2266, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 4.557602405548096, | |
| "learning_rate": 6.97737556561086e-05, | |
| "loss": 3.1531, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 3.354978354978355, | |
| "grad_norm": 4.508275508880615, | |
| "learning_rate": 6.886877828054299e-05, | |
| "loss": 3.2391, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.3766233766233764, | |
| "grad_norm": 4.455708980560303, | |
| "learning_rate": 6.796380090497739e-05, | |
| "loss": 3.1078, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 3.398268398268398, | |
| "grad_norm": 4.379371643066406, | |
| "learning_rate": 6.705882352941176e-05, | |
| "loss": 3.1875, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 3.41991341991342, | |
| "grad_norm": 5.268296241760254, | |
| "learning_rate": 6.615384615384616e-05, | |
| "loss": 3.1094, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 3.4415584415584415, | |
| "grad_norm": 3.684109687805176, | |
| "learning_rate": 6.524886877828055e-05, | |
| "loss": 3.2156, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 3.463203463203463, | |
| "grad_norm": 4.146853923797607, | |
| "learning_rate": 6.434389140271494e-05, | |
| "loss": 3.1953, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.463203463203463, | |
| "eval_loss": 2.1688125133514404, | |
| "eval_runtime": 6.4486, | |
| "eval_samples_per_second": 77.537, | |
| "eval_steps_per_second": 1.241, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.484848484848485, | |
| "grad_norm": 4.214244365692139, | |
| "learning_rate": 6.343891402714932e-05, | |
| "loss": 3.1297, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 3.5064935064935066, | |
| "grad_norm": 4.9321184158325195, | |
| "learning_rate": 6.253393665158372e-05, | |
| "loss": 3.1531, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 3.5281385281385282, | |
| "grad_norm": 4.7060017585754395, | |
| "learning_rate": 6.16289592760181e-05, | |
| "loss": 3.1703, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 3.54978354978355, | |
| "grad_norm": 4.779506206512451, | |
| "learning_rate": 6.072398190045249e-05, | |
| "loss": 3.1172, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 4.440573692321777, | |
| "learning_rate": 5.981900452488688e-05, | |
| "loss": 3.1437, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.5930735930735933, | |
| "grad_norm": 4.986171722412109, | |
| "learning_rate": 5.891402714932127e-05, | |
| "loss": 3.1938, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 3.6147186147186146, | |
| "grad_norm": 4.413197994232178, | |
| "learning_rate": 5.800904977375567e-05, | |
| "loss": 3.1422, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 4.145831108093262, | |
| "learning_rate": 5.7104072398190044e-05, | |
| "loss": 3.1297, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 3.658008658008658, | |
| "grad_norm": 4.501436710357666, | |
| "learning_rate": 5.6199095022624435e-05, | |
| "loss": 3.1172, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 3.6796536796536796, | |
| "grad_norm": 4.262230396270752, | |
| "learning_rate": 5.529411764705883e-05, | |
| "loss": 3.1875, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.6796536796536796, | |
| "eval_loss": 2.1631250381469727, | |
| "eval_runtime": 7.0489, | |
| "eval_samples_per_second": 70.933, | |
| "eval_steps_per_second": 1.135, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.7012987012987013, | |
| "grad_norm": 5.456052780151367, | |
| "learning_rate": 5.438914027149321e-05, | |
| "loss": 3.2078, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 3.722943722943723, | |
| "grad_norm": 4.408057689666748, | |
| "learning_rate": 5.34841628959276e-05, | |
| "loss": 3.1969, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 3.7445887445887447, | |
| "grad_norm": 3.87892484664917, | |
| "learning_rate": 5.2579185520362e-05, | |
| "loss": 3.1875, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 3.7662337662337664, | |
| "grad_norm": 4.127628326416016, | |
| "learning_rate": 5.167420814479639e-05, | |
| "loss": 3.1812, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 3.787878787878788, | |
| "grad_norm": 4.400271892547607, | |
| "learning_rate": 5.0769230769230766e-05, | |
| "loss": 3.1516, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.8095238095238093, | |
| "grad_norm": 4.7595601081848145, | |
| "learning_rate": 4.986425339366516e-05, | |
| "loss": 3.1484, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 3.8311688311688314, | |
| "grad_norm": 4.606581687927246, | |
| "learning_rate": 4.895927601809955e-05, | |
| "loss": 3.1375, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 3.8528138528138527, | |
| "grad_norm": 3.985682964324951, | |
| "learning_rate": 4.805429864253394e-05, | |
| "loss": 3.2391, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 3.8744588744588744, | |
| "grad_norm": 4.837043285369873, | |
| "learning_rate": 4.714932126696833e-05, | |
| "loss": 3.1375, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 3.896103896103896, | |
| "grad_norm": 5.133884429931641, | |
| "learning_rate": 4.624434389140272e-05, | |
| "loss": 3.1984, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.896103896103896, | |
| "eval_loss": 2.1586875915527344, | |
| "eval_runtime": 6.4439, | |
| "eval_samples_per_second": 77.593, | |
| "eval_steps_per_second": 1.241, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.9177489177489178, | |
| "grad_norm": 4.777352333068848, | |
| "learning_rate": 4.5339366515837104e-05, | |
| "loss": 3.0766, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 3.9393939393939394, | |
| "grad_norm": 4.657353401184082, | |
| "learning_rate": 4.4434389140271495e-05, | |
| "loss": 3.0906, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 3.961038961038961, | |
| "grad_norm": 4.391796588897705, | |
| "learning_rate": 4.3529411764705885e-05, | |
| "loss": 3.1688, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 3.982683982683983, | |
| "grad_norm": 4.547649383544922, | |
| "learning_rate": 4.262443438914027e-05, | |
| "loss": 3.1562, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 4.004329004329004, | |
| "grad_norm": 4.718013286590576, | |
| "learning_rate": 4.171945701357467e-05, | |
| "loss": 3.1313, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.025974025974026, | |
| "grad_norm": 5.101125240325928, | |
| "learning_rate": 4.081447963800905e-05, | |
| "loss": 3.0547, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 4.0476190476190474, | |
| "grad_norm": 4.414271831512451, | |
| "learning_rate": 3.990950226244344e-05, | |
| "loss": 3.0719, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 4.06926406926407, | |
| "grad_norm": 3.9164297580718994, | |
| "learning_rate": 3.900452488687783e-05, | |
| "loss": 3.1375, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 4.090909090909091, | |
| "grad_norm": 5.559360027313232, | |
| "learning_rate": 3.8099547511312216e-05, | |
| "loss": 3.1094, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 4.112554112554113, | |
| "grad_norm": 4.572648525238037, | |
| "learning_rate": 3.719457013574661e-05, | |
| "loss": 3.0641, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.112554112554113, | |
| "eval_loss": 2.1291251182556152, | |
| "eval_runtime": 6.5526, | |
| "eval_samples_per_second": 76.306, | |
| "eval_steps_per_second": 1.221, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.134199134199134, | |
| "grad_norm": 4.143476486206055, | |
| "learning_rate": 3.6289592760181e-05, | |
| "loss": 3.1281, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 4.1558441558441555, | |
| "grad_norm": 5.0819525718688965, | |
| "learning_rate": 3.538461538461539e-05, | |
| "loss": 3.1375, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 4.177489177489178, | |
| "grad_norm": 4.748097896575928, | |
| "learning_rate": 3.447963800904977e-05, | |
| "loss": 3.075, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 4.199134199134199, | |
| "grad_norm": 4.316249847412109, | |
| "learning_rate": 3.3574660633484164e-05, | |
| "loss": 2.9859, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 4.220779220779221, | |
| "grad_norm": 4.169147491455078, | |
| "learning_rate": 3.2669683257918554e-05, | |
| "loss": 3.1203, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.242424242424242, | |
| "grad_norm": 4.865535259246826, | |
| "learning_rate": 3.176470588235294e-05, | |
| "loss": 3.05, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 4.264069264069264, | |
| "grad_norm": 4.354024410247803, | |
| "learning_rate": 3.0859728506787336e-05, | |
| "loss": 3.0828, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 4.201045989990234, | |
| "learning_rate": 2.995475113122172e-05, | |
| "loss": 3.1562, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 4.307359307359308, | |
| "grad_norm": 4.673395156860352, | |
| "learning_rate": 2.904977375565611e-05, | |
| "loss": 3.0219, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "grad_norm": 4.373682975769043, | |
| "learning_rate": 2.8144796380090498e-05, | |
| "loss": 3.1125, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "eval_loss": 2.137437582015991, | |
| "eval_runtime": 6.3906, | |
| "eval_samples_per_second": 78.24, | |
| "eval_steps_per_second": 1.252, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.35064935064935, | |
| "grad_norm": 4.382651329040527, | |
| "learning_rate": 2.7239819004524886e-05, | |
| "loss": 3.0234, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 4.372294372294372, | |
| "grad_norm": 4.836797714233398, | |
| "learning_rate": 2.633484162895928e-05, | |
| "loss": 3.1156, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 4.393939393939394, | |
| "grad_norm": 4.31813907623291, | |
| "learning_rate": 2.5429864253393664e-05, | |
| "loss": 3.0484, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 4.415584415584416, | |
| "grad_norm": 4.9742350578308105, | |
| "learning_rate": 2.4524886877828054e-05, | |
| "loss": 3.0297, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 4.437229437229437, | |
| "grad_norm": 4.34569787979126, | |
| "learning_rate": 2.3619909502262445e-05, | |
| "loss": 3.0688, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 4.458874458874459, | |
| "grad_norm": 4.9268798828125, | |
| "learning_rate": 2.2714932126696833e-05, | |
| "loss": 3.0797, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 4.48051948051948, | |
| "grad_norm": 4.407137870788574, | |
| "learning_rate": 2.1809954751131223e-05, | |
| "loss": 3.0266, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 4.5021645021645025, | |
| "grad_norm": 4.480093479156494, | |
| "learning_rate": 2.090497737556561e-05, | |
| "loss": 3.0688, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 4.523809523809524, | |
| "grad_norm": 5.038318634033203, | |
| "learning_rate": 2e-05, | |
| "loss": 3.125, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 4.633235454559326, | |
| "learning_rate": 1.909502262443439e-05, | |
| "loss": 3.0562, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "eval_loss": 2.1156249046325684, | |
| "eval_runtime": 6.6798, | |
| "eval_samples_per_second": 74.853, | |
| "eval_steps_per_second": 1.198, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.567099567099567, | |
| "grad_norm": 5.131459712982178, | |
| "learning_rate": 1.819004524886878e-05, | |
| "loss": 3.1156, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 4.588744588744589, | |
| "grad_norm": 4.459514141082764, | |
| "learning_rate": 1.7285067873303167e-05, | |
| "loss": 3.0562, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 4.6103896103896105, | |
| "grad_norm": 4.892005443572998, | |
| "learning_rate": 1.6380090497737558e-05, | |
| "loss": 3.0297, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 4.632034632034632, | |
| "grad_norm": 4.963741779327393, | |
| "learning_rate": 1.5475113122171945e-05, | |
| "loss": 3.0125, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 4.653679653679654, | |
| "grad_norm": 4.354393482208252, | |
| "learning_rate": 1.4570135746606336e-05, | |
| "loss": 3.0578, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.675324675324675, | |
| "grad_norm": 4.8484649658203125, | |
| "learning_rate": 1.3665158371040724e-05, | |
| "loss": 3.0953, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 4.696969696969697, | |
| "grad_norm": 5.201778411865234, | |
| "learning_rate": 1.2760180995475113e-05, | |
| "loss": 3.0828, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 4.7186147186147185, | |
| "grad_norm": 5.245253086090088, | |
| "learning_rate": 1.1855203619909502e-05, | |
| "loss": 2.9922, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 4.740259740259741, | |
| "grad_norm": 5.066870212554932, | |
| "learning_rate": 1.0950226244343893e-05, | |
| "loss": 3.0516, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "grad_norm": 4.195698261260986, | |
| "learning_rate": 1.0045248868778282e-05, | |
| "loss": 3.0734, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.761904761904762, | |
| "eval_loss": 2.101749897003174, | |
| "eval_runtime": 6.5232, | |
| "eval_samples_per_second": 76.65, | |
| "eval_steps_per_second": 1.226, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.783549783549784, | |
| "grad_norm": 4.493175506591797, | |
| "learning_rate": 9.140271493212669e-06, | |
| "loss": 3.0187, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 4.805194805194805, | |
| "grad_norm": 4.933455944061279, | |
| "learning_rate": 8.23529411764706e-06, | |
| "loss": 3.0734, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 4.8268398268398265, | |
| "grad_norm": 4.937650203704834, | |
| "learning_rate": 7.330316742081449e-06, | |
| "loss": 3.0609, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 4.848484848484849, | |
| "grad_norm": 4.216976165771484, | |
| "learning_rate": 6.425339366515837e-06, | |
| "loss": 3.0156, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 4.87012987012987, | |
| "grad_norm": 4.135624408721924, | |
| "learning_rate": 5.520361990950226e-06, | |
| "loss": 3.0109, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.891774891774892, | |
| "grad_norm": 5.964848041534424, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 2.9969, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 4.913419913419913, | |
| "grad_norm": 4.022227764129639, | |
| "learning_rate": 3.710407239819005e-06, | |
| "loss": 3.0078, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 4.935064935064935, | |
| "grad_norm": 4.731805801391602, | |
| "learning_rate": 2.805429864253394e-06, | |
| "loss": 2.9844, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 4.956709956709957, | |
| "grad_norm": 4.872863292694092, | |
| "learning_rate": 1.9004524886877828e-06, | |
| "loss": 3.0484, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 4.978354978354979, | |
| "grad_norm": 4.713351249694824, | |
| "learning_rate": 9.954751131221719e-07, | |
| "loss": 3.0469, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.978354978354979, | |
| "eval_loss": 2.1040000915527344, | |
| "eval_runtime": 6.4203, | |
| "eval_samples_per_second": 77.878, | |
| "eval_steps_per_second": 1.246, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.077022075653076, | |
| "learning_rate": 9.04977375565611e-08, | |
| "loss": 3.0016, | |
| "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": 0.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |