| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.92436974789916, | |
| "eval_steps": 8, | |
| "global_step": 87, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03361344537815126, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 2e-05, | |
| "loss": 2.6905, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03361344537815126, | |
| "eval_loss": 2.729154109954834, | |
| "eval_runtime": 20.6566, | |
| "eval_samples_per_second": 2.421, | |
| "eval_steps_per_second": 0.339, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06722689075630252, | |
| "grad_norm": 1.0390625, | |
| "learning_rate": 4e-05, | |
| "loss": 2.687, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10084033613445378, | |
| "grad_norm": 1.1171875, | |
| "learning_rate": 6e-05, | |
| "loss": 2.8062, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13445378151260504, | |
| "grad_norm": 1.03125, | |
| "learning_rate": 8e-05, | |
| "loss": 2.6619, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16806722689075632, | |
| "grad_norm": 1.0, | |
| "learning_rate": 0.0001, | |
| "loss": 2.5939, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.20168067226890757, | |
| "grad_norm": 0.95703125, | |
| "learning_rate": 0.00012, | |
| "loss": 2.6307, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 0.86328125, | |
| "learning_rate": 0.00014, | |
| "loss": 2.5751, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 0.00016, | |
| "loss": 2.4725, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "eval_loss": 2.3971643447875977, | |
| "eval_runtime": 20.4746, | |
| "eval_samples_per_second": 2.442, | |
| "eval_steps_per_second": 0.342, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3025210084033613, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 0.00018, | |
| "loss": 2.2905, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.33613445378151263, | |
| "grad_norm": 0.6015625, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3883, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3697478991596639, | |
| "grad_norm": 0.53125, | |
| "learning_rate": 0.00019995608365087946, | |
| "loss": 2.3071, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.40336134453781514, | |
| "grad_norm": 0.46484375, | |
| "learning_rate": 0.00019982437317643217, | |
| "loss": 2.0817, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4369747899159664, | |
| "grad_norm": 0.56640625, | |
| "learning_rate": 0.0001996049842615217, | |
| "loss": 2.0883, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 0.5, | |
| "learning_rate": 0.00019929810960135172, | |
| "loss": 2.1018, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5042016806722689, | |
| "grad_norm": 0.5390625, | |
| "learning_rate": 0.0001989040187322164, | |
| "loss": 2.1015, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.00019842305779475968, | |
| "loss": 1.9891, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "eval_loss": 2.071833610534668, | |
| "eval_runtime": 20.5312, | |
| "eval_samples_per_second": 2.435, | |
| "eval_steps_per_second": 0.341, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.4921875, | |
| "learning_rate": 0.0001978556492299504, | |
| "loss": 2.1245, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6050420168067226, | |
| "grad_norm": 0.46875, | |
| "learning_rate": 0.0001972022914080411, | |
| "loss": 1.9485, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6386554621848739, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 0.00019646355819083589, | |
| "loss": 2.0621, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6722689075630253, | |
| "grad_norm": 0.41015625, | |
| "learning_rate": 0.00019564009842765225, | |
| "loss": 1.9076, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 0.4375, | |
| "learning_rate": 0.00019473263538541914, | |
| "loss": 1.8968, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7394957983193278, | |
| "grad_norm": 0.326171875, | |
| "learning_rate": 0.0001937419661134121, | |
| "loss": 1.8977, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.773109243697479, | |
| "grad_norm": 0.314453125, | |
| "learning_rate": 0.00019266896074318334, | |
| "loss": 1.9926, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.00019151456172430183, | |
| "loss": 1.8345, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "eval_loss": 1.932895302772522, | |
| "eval_runtime": 20.5148, | |
| "eval_samples_per_second": 2.437, | |
| "eval_steps_per_second": 0.341, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "grad_norm": 0.291015625, | |
| "learning_rate": 0.00019027978299657436, | |
| "loss": 1.8168, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8739495798319328, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.00018896570909947475, | |
| "loss": 1.8794, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.907563025210084, | |
| "grad_norm": 0.306640625, | |
| "learning_rate": 0.0001875734942195637, | |
| "loss": 1.9604, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.00018610436117673555, | |
| "loss": 1.824, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9747899159663865, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.0001845596003501826, | |
| "loss": 1.9186, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.0084033613445378, | |
| "grad_norm": 0.40625, | |
| "learning_rate": 0.0001829405685450202, | |
| "loss": 1.9281, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0420168067226891, | |
| "grad_norm": 0.396484375, | |
| "learning_rate": 0.00018124868780056814, | |
| "loss": 1.833, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.0756302521008403, | |
| "grad_norm": 0.28515625, | |
| "learning_rate": 0.00017948544414133534, | |
| "loss": 1.8088, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.0756302521008403, | |
| "eval_loss": 1.8729643821716309, | |
| "eval_runtime": 20.5377, | |
| "eval_samples_per_second": 2.435, | |
| "eval_steps_per_second": 0.341, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.1092436974789917, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.00017765238627180424, | |
| "loss": 1.8401, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.2451171875, | |
| "learning_rate": 0.00017575112421616202, | |
| "loss": 1.7736, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.2099609375, | |
| "learning_rate": 0.00017378332790417273, | |
| "loss": 1.7875, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.2100840336134453, | |
| "grad_norm": 0.1767578125, | |
| "learning_rate": 0.00017175072570443312, | |
| "loss": 1.7743, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.2436974789915967, | |
| "grad_norm": 0.1982421875, | |
| "learning_rate": 0.00016965510290629972, | |
| "loss": 1.8167, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.2773109243697478, | |
| "grad_norm": 0.2119140625, | |
| "learning_rate": 0.00016749830015182107, | |
| "loss": 1.8317, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.3109243697478992, | |
| "grad_norm": 0.1796875, | |
| "learning_rate": 0.00016528221181905217, | |
| "loss": 1.7498, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.3445378151260505, | |
| "grad_norm": 0.197265625, | |
| "learning_rate": 0.00016300878435817113, | |
| "loss": 1.8183, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3445378151260505, | |
| "eval_loss": 1.8429957628250122, | |
| "eval_runtime": 20.5619, | |
| "eval_samples_per_second": 2.432, | |
| "eval_steps_per_second": 0.34, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3781512605042017, | |
| "grad_norm": 0.193359375, | |
| "learning_rate": 0.00016068001458185936, | |
| "loss": 1.7504, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 0.0001582979479114472, | |
| "loss": 1.7476, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.4453781512605042, | |
| "grad_norm": 0.1787109375, | |
| "learning_rate": 0.00015586467658036524, | |
| "loss": 1.8082, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.4789915966386555, | |
| "grad_norm": 0.1669921875, | |
| "learning_rate": 0.0001533823377964791, | |
| "loss": 1.6987, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.5126050420168067, | |
| "grad_norm": 0.1728515625, | |
| "learning_rate": 0.00015085311186492206, | |
| "loss": 1.7417, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.5462184873949578, | |
| "grad_norm": 0.1826171875, | |
| "learning_rate": 0.00014827922027307451, | |
| "loss": 1.779, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.5798319327731094, | |
| "grad_norm": 0.1826171875, | |
| "learning_rate": 0.0001456629237393713, | |
| "loss": 1.7552, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.6134453781512605, | |
| "grad_norm": 0.1767578125, | |
| "learning_rate": 0.00014300652022765207, | |
| "loss": 1.8004, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6134453781512605, | |
| "eval_loss": 1.8263332843780518, | |
| "eval_runtime": 20.4305, | |
| "eval_samples_per_second": 2.447, | |
| "eval_steps_per_second": 0.343, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 0.19140625, | |
| "learning_rate": 0.00014031234292879725, | |
| "loss": 1.7857, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.680672268907563, | |
| "grad_norm": 0.181640625, | |
| "learning_rate": 0.00013758275821142382, | |
| "loss": 1.7193, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.1611328125, | |
| "learning_rate": 0.0001348201635434399, | |
| "loss": 1.7734, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.7478991596638656, | |
| "grad_norm": 0.169921875, | |
| "learning_rate": 0.00013202698538628376, | |
| "loss": 1.7644, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.7815126050420167, | |
| "grad_norm": 0.171875, | |
| "learning_rate": 0.00012920567706369758, | |
| "loss": 1.8631, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.815126050420168, | |
| "grad_norm": 0.1845703125, | |
| "learning_rate": 0.00012635871660690676, | |
| "loss": 1.7735, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.8487394957983194, | |
| "grad_norm": 0.1689453125, | |
| "learning_rate": 0.00012348860457809838, | |
| "loss": 1.7956, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 0.1708984375, | |
| "learning_rate": 0.00012059786187410984, | |
| "loss": 1.7674, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "eval_loss": 1.8166663646697998, | |
| "eval_runtime": 20.5178, | |
| "eval_samples_per_second": 2.437, | |
| "eval_steps_per_second": 0.341, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.9159663865546217, | |
| "grad_norm": 0.169921875, | |
| "learning_rate": 0.0001176890275122573, | |
| "loss": 1.7228, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.949579831932773, | |
| "grad_norm": 0.169921875, | |
| "learning_rate": 0.00011476465640024814, | |
| "loss": 1.7496, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.9831932773109244, | |
| "grad_norm": 0.173828125, | |
| "learning_rate": 0.00011182731709213659, | |
| "loss": 1.6776, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.0168067226890756, | |
| "grad_norm": 0.171875, | |
| "learning_rate": 0.00010887958953229349, | |
| "loss": 1.7346, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0504201680672267, | |
| "grad_norm": 0.1669921875, | |
| "learning_rate": 0.00010592406278937144, | |
| "loss": 1.8371, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.0840336134453783, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 0.00010296333278225599, | |
| "loss": 1.7113, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 0.14453125, | |
| "learning_rate": 0.0001, | |
| "loss": 1.7299, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.1512605042016806, | |
| "grad_norm": 0.154296875, | |
| "learning_rate": 9.703666721774402e-05, | |
| "loss": 1.7164, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.1512605042016806, | |
| "eval_loss": 1.810401439666748, | |
| "eval_runtime": 20.5114, | |
| "eval_samples_per_second": 2.438, | |
| "eval_steps_per_second": 0.341, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.184873949579832, | |
| "grad_norm": 0.1572265625, | |
| "learning_rate": 9.407593721062859e-05, | |
| "loss": 1.746, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.2184873949579833, | |
| "grad_norm": 0.169921875, | |
| "learning_rate": 9.112041046770653e-05, | |
| "loss": 1.7507, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.2521008403361344, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 8.817268290786343e-05, | |
| "loss": 1.6836, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.154296875, | |
| "learning_rate": 8.523534359975189e-05, | |
| "loss": 1.7432, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.3193277310924367, | |
| "grad_norm": 0.16796875, | |
| "learning_rate": 8.231097248774274e-05, | |
| "loss": 1.7734, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.1591796875, | |
| "learning_rate": 7.940213812589018e-05, | |
| "loss": 1.6683, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.3865546218487395, | |
| "grad_norm": 0.154296875, | |
| "learning_rate": 7.651139542190164e-05, | |
| "loss": 1.6743, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.4201680672268906, | |
| "grad_norm": 0.1708984375, | |
| "learning_rate": 7.364128339309326e-05, | |
| "loss": 1.6525, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.4201680672268906, | |
| "eval_loss": 1.806851863861084, | |
| "eval_runtime": 20.4653, | |
| "eval_samples_per_second": 2.443, | |
| "eval_steps_per_second": 0.342, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.453781512605042, | |
| "grad_norm": 0.16015625, | |
| "learning_rate": 7.079432293630244e-05, | |
| "loss": 1.708, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.4873949579831933, | |
| "grad_norm": 0.16015625, | |
| "learning_rate": 6.797301461371625e-05, | |
| "loss": 1.6694, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.5210084033613445, | |
| "grad_norm": 0.162109375, | |
| "learning_rate": 6.517983645656014e-05, | |
| "loss": 1.5762, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.5546218487394956, | |
| "grad_norm": 0.1640625, | |
| "learning_rate": 6.24172417885762e-05, | |
| "loss": 1.7879, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.1591796875, | |
| "learning_rate": 5.96876570712028e-05, | |
| "loss": 1.6566, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.6218487394957983, | |
| "grad_norm": 0.16015625, | |
| "learning_rate": 5.699347977234799e-05, | |
| "loss": 1.686, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.6554621848739495, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 5.43370762606287e-05, | |
| "loss": 1.7537, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.689075630252101, | |
| "grad_norm": 0.162109375, | |
| "learning_rate": 5.172077972692553e-05, | |
| "loss": 1.7917, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.689075630252101, | |
| "eval_loss": 1.805298924446106, | |
| "eval_runtime": 20.4745, | |
| "eval_samples_per_second": 2.442, | |
| "eval_steps_per_second": 0.342, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.722689075630252, | |
| "grad_norm": 0.166015625, | |
| "learning_rate": 4.914688813507797e-05, | |
| "loss": 1.7791, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.7563025210084033, | |
| "grad_norm": 0.1748046875, | |
| "learning_rate": 4.661766220352097e-05, | |
| "loss": 1.7805, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.7899159663865545, | |
| "grad_norm": 0.1630859375, | |
| "learning_rate": 4.4135323419634766e-05, | |
| "loss": 1.6505, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.15625, | |
| "learning_rate": 4.170205208855281e-05, | |
| "loss": 1.6896, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.1630859375, | |
| "learning_rate": 3.931998541814069e-05, | |
| "loss": 1.6374, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.8907563025210083, | |
| "grad_norm": 0.1689453125, | |
| "learning_rate": 3.69912156418289e-05, | |
| "loss": 1.7444, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.92436974789916, | |
| "grad_norm": 0.1494140625, | |
| "learning_rate": 3.471778818094785e-05, | |
| "loss": 1.7065, | |
| "step": 87 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 29, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9933851341357056.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |